Month: April 2026
Call a Matlab function in Simulink for later code generation
Hello!
I need to use a MATLAB function in my Simulink project.
The problem is that this function needs to call other functions and load data from *.mat files (see image).
I also need to generate the code using Simulink Coder and Embedded Coder.
Which technique should I adopt from the answer below ?
Call a Matlab function in Simulink from the current folderHello!
I need to use a MATLAB function in my Simulink project.
The problem is that this function needs to call other functions and load data from *.mat files (see image).
I also need to generate the code using Simulink Coder and Embedded Coder.
Which technique should I adopt from the answer below ?
Call a Matlab function in Simulink from the current folder Hello!
I need to use a MATLAB function in my Simulink project.
The problem is that this function needs to call other functions and load data from *.mat files (see image).
I also need to generate the code using Simulink Coder and Embedded Coder.
Which technique should I adopt from the answer below ?
Call a Matlab function in Simulink from the current folder simulink, matlab, code generation, embedded matlab function, embedded coder MATLAB Answers — New Questions
Matlab online docs showing in dark mode
The Matlab online docs (e.g., https://www.mathworks.com/help/?s_tid=user_nav_help ) showing in dark mode on mathworks.com are showing in dark mode for me. Is there a site setting to switch them back to light mode?The Matlab online docs (e.g., https://www.mathworks.com/help/?s_tid=user_nav_help ) showing in dark mode on mathworks.com are showing in dark mode for me. Is there a site setting to switch them back to light mode? The Matlab online docs (e.g., https://www.mathworks.com/help/?s_tid=user_nav_help ) showing in dark mode on mathworks.com are showing in dark mode for me. Is there a site setting to switch them back to light mode? matlab MATLAB Answers — New Questions
comm.SDRuReceiver on X310 throws receiveData:ErrLateCommand
I am trying to receive data from a USRP X310 in MATLAB using comm.SDRuReceiver, and I keep hitting the error:
Receive unsuccessfully: Could not execute UHD driver command in ‘receiveData_cont_c’: receiveData:ErrLateCommand. A stream command was issued in the past.
Running the benchmark_rate reports: * 0 dropped samples * 0 overflows * 0 late commands * 0 timeoutsI am trying to receive data from a USRP X310 in MATLAB using comm.SDRuReceiver, and I keep hitting the error:
Receive unsuccessfully: Could not execute UHD driver command in ‘receiveData_cont_c’: receiveData:ErrLateCommand. A stream command was issued in the past.
Running the benchmark_rate reports: * 0 dropped samples * 0 overflows * 0 late commands * 0 timeouts I am trying to receive data from a USRP X310 in MATLAB using comm.SDRuReceiver, and I keep hitting the error:
Receive unsuccessfully: Could not execute UHD driver command in ‘receiveData_cont_c’: receiveData:ErrLateCommand. A stream command was issued in the past.
Running the benchmark_rate reports: * 0 dropped samples * 0 overflows * 0 late commands * 0 timeouts usrp, comm.sdrureceiver, x310 MATLAB Answers — New Questions
How to find a function in the path – securely?
What is a safe method to check, where a specific function exists in Matlab’s path? It should not matter, if the function is an M, P or MEX file.
Actually, this is a job for which() . To avoid collisions with local variables, it should be hidden in a function:
function Reply = safeWhich(varargin)
Reply = which(varargin{:});
end
This fails for ‘varargin’, as expected:
safeWhich(‘varargin’)
But there are further traps
safeWhich(‘R’) % Same for ‘P’
In Matlab R2018b under Windows I get:
‘C:Program FilesMATLABR2018btoolboxmatlabcodetools@mtreemtree.m % mtree method’
The appended comment allowed to exclude the output by checking, if a corrensponding file exists:
function Reply = safeWhich_2(varargin)
Reply = which(varargin{:});
Reply = cellstr(Reply);
Reply = Reply(isfile(Reply)); % Worked in R2018b, no effect in R2025a
end
safeWhich_2(‘R’)
The exist() command does have the power to find functions properly, but it does not reveal where.
The old function depfun() was replaced by the ugly matlab.codetools.requiredFilesAndProducts(). I could search in the corresponding code, couldn’t I? This function uses the code find here internally: "toolboxmatlabdepfun+matlab+depfun+internal+whichcallWhich.m". Here I find 91 functions and 11.3 MB of code. One function looks like this:
% 10^(-18) is effective zero in terms of possibility, which requries
% log(10^18)/log(26+10+1) = 12 random letters, digits, or underscores.
function klno6phn_9faskf_na = callWhich(asm_foyan_knaouie8)
klno6phn_9faskf_na = which(asm_foyan_knaouie8);
end
Obviously MathWorks struggles with comparable problems.
The profiler shows, that matlab.codetools.requiredFilesAndProducts() calls 207 subfunctions (R2018b).
See the discussion: https://www.mathworks.com/matlabcentral/discussions/ideas/887763-i-have-been-a-matlab-loyalist-for-25-years . The complexity explodes, such that a simple task like searching a function file needs a pile of exceptions and indirections.What is a safe method to check, where a specific function exists in Matlab’s path? It should not matter, if the function is an M, P or MEX file.
Actually, this is a job for which() . To avoid collisions with local variables, it should be hidden in a function:
function Reply = safeWhich(varargin)
Reply = which(varargin{:});
end
This fails for ‘varargin’, as expected:
safeWhich(‘varargin’)
But there are further traps
safeWhich(‘R’) % Same for ‘P’
In Matlab R2018b under Windows I get:
‘C:Program FilesMATLABR2018btoolboxmatlabcodetools@mtreemtree.m % mtree method’
The appended comment allowed to exclude the output by checking, if a corrensponding file exists:
function Reply = safeWhich_2(varargin)
Reply = which(varargin{:});
Reply = cellstr(Reply);
Reply = Reply(isfile(Reply)); % Worked in R2018b, no effect in R2025a
end
safeWhich_2(‘R’)
The exist() command does have the power to find functions properly, but it does not reveal where.
The old function depfun() was replaced by the ugly matlab.codetools.requiredFilesAndProducts(). I could search in the corresponding code, couldn’t I? This function uses the code find here internally: "toolboxmatlabdepfun+matlab+depfun+internal+whichcallWhich.m". Here I find 91 functions and 11.3 MB of code. One function looks like this:
% 10^(-18) is effective zero in terms of possibility, which requries
% log(10^18)/log(26+10+1) = 12 random letters, digits, or underscores.
function klno6phn_9faskf_na = callWhich(asm_foyan_knaouie8)
klno6phn_9faskf_na = which(asm_foyan_knaouie8);
end
Obviously MathWorks struggles with comparable problems.
The profiler shows, that matlab.codetools.requiredFilesAndProducts() calls 207 subfunctions (R2018b).
See the discussion: https://www.mathworks.com/matlabcentral/discussions/ideas/887763-i-have-been-a-matlab-loyalist-for-25-years . The complexity explodes, such that a simple task like searching a function file needs a pile of exceptions and indirections. What is a safe method to check, where a specific function exists in Matlab’s path? It should not matter, if the function is an M, P or MEX file.
Actually, this is a job for which() . To avoid collisions with local variables, it should be hidden in a function:
function Reply = safeWhich(varargin)
Reply = which(varargin{:});
end
This fails for ‘varargin’, as expected:
safeWhich(‘varargin’)
But there are further traps
safeWhich(‘R’) % Same for ‘P’
In Matlab R2018b under Windows I get:
‘C:Program FilesMATLABR2018btoolboxmatlabcodetools@mtreemtree.m % mtree method’
The appended comment allowed to exclude the output by checking, if a corrensponding file exists:
function Reply = safeWhich_2(varargin)
Reply = which(varargin{:});
Reply = cellstr(Reply);
Reply = Reply(isfile(Reply)); % Worked in R2018b, no effect in R2025a
end
safeWhich_2(‘R’)
The exist() command does have the power to find functions properly, but it does not reveal where.
The old function depfun() was replaced by the ugly matlab.codetools.requiredFilesAndProducts(). I could search in the corresponding code, couldn’t I? This function uses the code find here internally: "toolboxmatlabdepfun+matlab+depfun+internal+whichcallWhich.m". Here I find 91 functions and 11.3 MB of code. One function looks like this:
% 10^(-18) is effective zero in terms of possibility, which requries
% log(10^18)/log(26+10+1) = 12 random letters, digits, or underscores.
function klno6phn_9faskf_na = callWhich(asm_foyan_knaouie8)
klno6phn_9faskf_na = which(asm_foyan_knaouie8);
end
Obviously MathWorks struggles with comparable problems.
The profiler shows, that matlab.codetools.requiredFilesAndProducts() calls 207 subfunctions (R2018b).
See the discussion: https://www.mathworks.com/matlabcentral/discussions/ideas/887763-i-have-been-a-matlab-loyalist-for-25-years . The complexity explodes, such that a simple task like searching a function file needs a pile of exceptions and indirections. which, exist, required products, file MATLAB Answers — New Questions
High Volume Email is Generally Available and Ready to Charge
Pricing for HVE Announced Effective June 1, 2026
In March 2026, Microsoft announced that general availability of the Exchange Online High-Volume Email (HVE) solution was imminent. On April 1, 2026, HVE duly attained general availability in a Technical Community post that revealed details of pricing amongst other things.
Paying to Send to Internal Recipients
The headline rate for HVE is $42 (U.S.) per one million email recipients, or $0.000042 per recipient. All recipients are internal because HVE doesn’t support sending email outside the tenant.
At first glance, the pricing is much lower than Azure Email Communication Service (ECS), which quotes $274 to send one million messages to external recipients. However, this is not an apple-to-apple comparison because one charges on a per-recipient basis while the other charges on a per-message basis. It’s entirely possible that each message will have a single recipient, but that’s not always the case.
ECS cites an average email size of 0.2 MB for the estimate cited above. The larger the message, the more you’ll pay. Increasing the size to 1 MB moves the dial to $370 according to the Azure price calculator. By comparison, HVE sets an upper limit of 10 MB per message.
Sending email with ECS isn’t difficult (here’s an example of using ECS to send an Outlook newsletter), if you have a moderate acquaintance with PowerShell. ECS comes with features that HVE doesn’t have, such as tracking message delivery. You can argue that this kind of feature isn’t necessary for an internally-facing email service because tools like message tracing for internal recipients are available out-of-the-box.
Configure a Billing Policy
After June 1, 2026, Microsoft will charge for HVE on a pay as you go (PAYG) basis through a credit card attached to an Azure subscription. To make sure that you can keep on using HVE after Microsoft enables charging, you’ll need to create a billing policy through the Pay as you go option under Billing in the Microsoft 365 admin center. Creating a billing policy essentially links a valid (active) Azure subscription with a PAYG service. During setup, you should create a new Azure resource group for HVE processing (Figure 1) because this will make Azure billing easier to understand. You can choose an existing resource group if you wish.
During setup, you get a chance to set a budget for the billing policy. This is to make sure that the PAYG service doesn’t exceed the expected level of cost during the monthly billing period. You can configure a mail-enabled security group to receive notifications when a budget percentage threshold is reached (Figure 2).
The final step is to assign the billing policy to the HVE service (Figure 3). Once connected, HVE will be ready to meter processing and generate charges to the credit card connected through the policy when Microsoft enables billing for HVE on June 1, 2026.
Billing Policy Management via PowerShell
PowerShell cmdlets are available to view and manage billing policies (see the Microsoft documentation for managing HVE). I don’t imagine that many will use PowerShell to manage billing policies because the admin UX works well, but here’s an example of running the Get-BillingPolicy cmdlet to view the details of the policy created above.
Get-BillingPolicy -ResourceType HVE
Id : c3601166-04c3-407b-9cf9-af7a5fa70bfe
FriendlyName : HVE Email Policy
IdentityType : Tenant
IdentityId : a662313f-14fc-43a2-9a7a-d2e27f4f3478
FeatureSet : {HighVolumeEmail}
TenantId : b662313f-14fc-43a2-9a7a-d2e27f4f3478
SubscriptionId : 35429342-a1a5-4427-9e2d-551840f2ad25
SubscriptionState : Registered
ResourceGroupName : HVE
ResourceProviderName : Microsoft.Syntex/accounts
AzureRegion : West Europe
ResourceUri : /subscriptions/35429342-a1a5-4427-9e2d-551840f2ad25/resourceGroups/HVE/providers/Microsoft.Syntex/accounts/c3601166-04c3-407b-9cf9-af7a5fa70bfe
CreateDate : 21/04/2026 20:48:53
LastModifiedDate : 21/04/2026 20:50:42
What’s interesting here is that the resource provider name is Microsoft.Syntex, usually something that’s associated with SharePoint Online. This is likely explained by Exchange Online adopting the same PAYG billing mechanism as used by SharePoint services such as document translation.
All the cmdlets are in the Exchange Online management PowerShell module.
Setting a Reply-to Address for HVE Accounts
One thing that you will want to do with PowerShell is create a reply-to address for HVE accounts. This is a recent improvement to eliminate the problem of how to handle responses to HVE messages. Until recently, if a recipient replied to a message, the response when into a black hole because HVE accounts don’t have mailboxes. By configuring the accounts with a reply-to address, Exchange Online can redirect messages to a “real” address. This code shows what I did to find all HVE accounts in the tenant and set the same address (for a shared mailbox) for each.
[array]$HVEAccounts = Get-MailUser -HVEAccount
ForEach ($Account in $HVEAccounts) {
Set-HVEAccountSettings -Identity $Account.WindowsEmailAddress -ReplyTo "cServices@office365itpros.com"
}
Two Bulk Email Solutions When One Will Do
Now that HVE is generally available, Microsoft 365 tenants have a choice of two email platforms for bulk email (well, hybrid tenants can continue using Exchange Server and forget about the cloud platforms). Both HVE and ECS are built on Exchange Online. Both charge to send email. Wouldn’t it make a heap of sense to create just one bulk email platform that can handle both internal and external traffic? And while creating a unified bulk email platform, make sure that it’s much easier to use than either solution is today, especially for hardware devices like printers.
Organizational politics within Microsoft and available budgets might get in the way of creating a single unified bulk email platform built on Exchange Online, but it’s the right long-term direction.
Insight like this doesn’t come easily. You’ve got to know the technology and understand how to look behind the scenes. Benefit from the knowledge and experience of the Office 365 for IT Pros team by subscribing to the best eBook covering Office 365 and the wider Microsoft 365 ecosystem.
Accelerating Frontier Transformation with Microsoft partners
AI has moved quickly from experimentation to production. Customers want measurable business outcomes, along with security, governance and responsible AI built in from day one. Microsoft partners are a meaningful differentiator to deliver these objectives. They turn ideas into deployable solutions by prioritizing the highest value use cases, building the right data and security foundations and establishing adoption and measurement capabilities so customers can run AI reliably in production.
Frontier Transformation is where AI becomes a repeatable, governed capability embedded into the flow of work, business processes and customer engagement. Customers are quickly moving from targeted pilots to operating AI at scale with a foundation built upon identity, data protection, compliance, monitoring and change management. As organizations expand from custom agents to agent-led processes, unified governance is essential so leaders can manage risk, track performance and scale with confidence.
Two essentials: Intelligence and Trust
Frontier Transformation depends on two essential elements: intelligence and trust. Customers want solutions grounded in their unique work intelligence, including their data, business context and operational realities. They also expect trust by design, with AI artifacts observable, managed and secured across the technology stack so they can deploy responsibly and scale with confidence.
A success framework for Frontier Transformation
Microsoft has developed a powerful framework for success as partners enable AI transformation for customers across all segments, industries and geographies:
- Enriching employee experiences: enabling businesses to empower employees with world-class tools and capabilities to activate a thriving, productive workforce
- Reinventing customer engagement: applying AI and agentic solutions to break through with customers, accelerate revenue growth, become more efficient at customer acquisition and deliver more personalized solutions
- Reshaping business processes: redesigning workflows across the business, enhanced by AI and agentic capability
- Bending the curve on innovation: AI acceleration is a powerful catalyst for business transformation and for addressing society’s biggest challenges — curing disease, addressing climate change and famine and other meaningful advancements
The “what” matters, and so does the “how.” Organizations that scale successfully put AI where people already work, enable innovation close to the business challenge and build observability at every layer so leaders can measure quality, govern risk and manage AI like a production system.
More than 90% of the Fortune 500 use Microsoft 365 Copilot, reflecting how quickly AI is becoming part of everyday work.(1) IDC predicts 1.3 billion agents in circulation by 2028(2) and 80% of the Fortune 500 are already using Microsoft agents, led by operationally complex industries like manufacturing, financial services and retail.(3) As customers move from piloting AI to agents embedded in their flow of work, governance and security need to scale with them.
Microsoft’s approach is straightforward: Copilot drives action in the flow of work, agents orchestrate workflows across systems and Microsoft Agent 365 provides a unified control plane designed to govern and secure agents at scale, with the same tools businesses use for employee administration, such as Microsoft admin center, Defender, Entra and Purview.
Partners are creating impact right now in three areas. First, agentic workflows that remove operational friction and orchestrate end-to-end work across operations, finance, supply chain and service. Second, Customer Zero maturity. Partners who adopt Copilot and agents internally build credibility and move faster because they have meaningful, real-world experiences that they translate into their go-to-market plans. Third, security as the foundation. There is no AI at scale without secure identity, protected data and strong governance.
Microsoft 365 E7 and Agent 365: The Frontier Suite
In March, Microsoft introduced Wave 3 of Microsoft 365 Copilot and announced Microsoft 365 E7: The Frontier Suite, with general availability of Microsoft 365 E7 and Microsoft Agent 365 on May 1, 2026.
Microsoft 365 E7 brings together Microsoft 365 E5 for secure productivity, Entra Suite for identity and access control, Microsoft 365 Copilot for AI in the flow of work and Agent 365 as the control plane to govern and scale agents. It is grounded in shared intelligence from Work IQ, the layer that brings together signals from the Microsoft 365 environment, including content, context and activity, so AI can operate with the right business grounding and policy awareness.
Microsoft Agent 365 provides a unified control plane for agents, enabling IT, security and business teams to observe, govern and secure agents across the organization. This applies to any agents an organization uses, whether they are built on Microsoft AI platforms, delivered by ecosystem partners or introduced through other technology stacks. It also applies the same security and compliance capabilities teams already rely on, including Microsoft Defender, Microsoft Entra and Microsoft Purview.
Some customer scenarios require custom agents. Microsoft Agent Factory is designed to accelerate the move from experimentation to execution. The Microsoft Agent Factory Pre-purchase Plan (P3) adds licensing flexibility across Copilot Studio, Microsoft Foundry, Fabric and GitHub, with tiered discounts intended to support broader adoption rather than isolated pilots. It also enables inclusion of tailored, role-based skilling at no additional cost to the customer, reducing adoption friction and increasing delivered value.
The opportunity for partners is end-to-end, and this is where the partner’s strategy really matters. Shifting from transaction-first to outcome-first, partners who iterate quickly, establish clear guardrails and build an operating rhythm for adoption move customers from interest to impact.
Over time, every organization will employ people who can direct and govern agents as part of daily work. Partners can make that capability real through packaged offers, change management and managed operations. Publishing those packaged offers in the Microsoft Marketplace adds a scalable route to market, improving discoverability and enabling a more repeatable buy-and-deploy motion as customers expand agent usage.
Partner success: What governed scale looks like in practice
“AI is at the forefront of everything we do. Through our ‘learn, use, create’ methodology and our AI Academy, we really support partners with learning paths.”
— Nicole Clark, Global Alliance Manager, Arrow Electronics
Partners are embracing Frontier Transformation by modernizing foundations, driving adoption, designing security into delivery and building agents that automate repeatable work and orchestrate business processes.
- Cognizant treated legacy automation as a platform modernization effort. Using Microsoft Power Platform, Copilot agents and governance frameworks, Cognizant migrated and modernized automation and scaled it across teams, consolidating platforms, lowering costs and reducing manual work through agent-led workflows.
- EPAM’s work with their customer Albert Heijn demonstrates what agent-first execution looks like in frontline scenarios. By delivering an employee-facing virtual assistant inside the retailer’s staff app, EPAM supported scenarios like restocking, onboarding and faster access to product and inventory information, with enterprise governance and observability in mind.
- Insight’s Flight Academy shows what it looks like to treat adoption as a program, not an announcement. Through a structured approach, Insight enables teams to build AI fluency in daily work and reinforces usage with practical learning and internal momentum that can scale beyond early enthusiasts.
- aCloud demonstrated a repeatable security pattern with their customer Jurong Engineering Limited (JEL) by bringing together Microsoft Purview, Microsoft Sentinel, Microsoft Defender XDR and Microsoft Security Copilot, paired with co-design workshops and cross-team alignment to strengthen compliance readiness.
- Arrow Electronics showed how distributor-led enablement can accelerate partner execution by using ArrowSphere to streamline Cloud Solution Provider (CSP) lifecycle management and ArrowSphere Assistant to surface AI-driven insights for renewals, upsell opportunities and Copilot adoption, complemented by a security dashboard that strengthens posture visibility and supports trust-by-design conversations.
Find more stories of partners innovating and driving meaningful outcomes for customers with Microsoft technology.
This same disciplined approach is especially relevant in the small and medium business space (SMB), where Microsoft partners offer end-to-end capability through managed services offerings and solutions packaged into repeatable motions, tailored to this customer segment.
SMB momentum: Scaling work with Copilots and agents
As Microsoft 365 Copilot Business expands AI built for work to organizations with fewer than 300 users, SMBs have a practical path to adopt AI more broadly. CSP partners are well positioned to guide that journey with a motion that combines adoption, security and ongoing management.
New Omdia research illustrates that CSP is a durable growth model for partners. In a study of 267 CSP partners across 36 countries, 79% rated CSP authorization as good, very good or excellent, and 88% would recommend it to other partners.(4) Omdia also found that 60% of CSP partner revenue is now tied to value-added services, with licensing acting as the entry point to broader, services-led engagements.(6)
“We’re bringing customers resources that only a partner can deliver to them: our relationship with Microsoft, technical training and programs that push them further and faster to learn technologies like Microsoft Copilot Studio, Foundry and Fabric.”
— Chance Weaver, Global VP of AI Adoption, Pax8
SMB demand is also expanding. For CSP partners, the near-term opportunity is to standardize advancing Copilot and agents from conversation to consumption. Lead with a simple, repeatable motion: outcome selection, security baseline, deployment, adoption and optimization cadence. Renewal moments are often the easiest time to introduce change when paired with a clear business case and time-bound offers.
A simple, scalable approach is to roll out in stages:
- Deploy Microsoft 365 Copilot Business broadly, paired with a strong foundation of identity, data protection and compliance.
- Target high-propensity accounts with tools such as Microsoft CloudAscent and the AI Business Solutions & Security Insights dashboard to deepen adoption and standardize responsible prompting.
- Extend with agents to take on repeatable tasks and support key business processes, with governance and security built in.
Microsoft provides CSP partners with a powerful set of tools to combine licensing, lifecycle management and optimization into one customer relationship. Omdia notes that partners value operational advantages such as monthly billing flexibility and managing licenses through Partner Center for real-time provisioning and 24/7 license management. Partners can review the CSP incentives guide to understand the latest CSP incentives and how they map to an SMB motion.
Microsoft supports SMB-focused partners by combining product, security and go-to-market resources that make it easy to deliver a repeatable motion. That includes tools to assess readiness, prioritize the right use cases and track adoption over time, plus role-based skilling to build sales and technical and delivery confidence across Copilot, security and agents. For partners building managed services offerings, Microsoft Marketplace also provides a scalable route to market, improving discoverability and enabling customers to buy through familiar procurement paths, while Partner Center brings licensing and lifecycle management into the same operational flow.
Program momentum and updates
The Microsoft AI Cloud Partner Program continues to be the primary way we invest in partners as they build, sell and deliver cloud and AI solutions. Our focus is simple: enable partners to build capability, accelerate demand, differentiate in the market and scale repeatable delivery.
In February 2026, Microsoft introduced a wealth of expanded benefits updates across Copilot, security, Azure credits and go-to-market resources. These updates are designed to strengthen how partners run their business and accelerate the ability to take solutions to market. We continue to evolve partner benefits packages as a practical growth lever, combining product, support and advisory benefits so partners can invest with confidence.
To enable AI Transformation, Microsoft is introducing program updates and offers in the coming months. These updates are intended to enable partners, including services partners, channel partners and software companies, to build and deliver agents across the Frontier product stack.
- Differentiation via Frontier Partner specialization: The Frontier Badge is evolving into a Frontier Partner specialization. This specialization differentiates partners, including services partners and channel partners, who demonstrate capabilities to build or deliver agents across Microsoft’s Frontier product stack. It creates a clear way for customers and Microsoft field sales teams to identify partners with validated readiness for agentic AI scenarios.
- Updated Frontier Distributor designation: Microsoft is evolving the Frontier Distributor designation to reflect distributor capabilities that matter for scaling agentic AI across the channel. For partners, this will make it easier to identify distributors that can deliver repeatable skilling, enablement to build and manage agents and Marketplace-backed motions to transact and grow agent sales.
- Benefits for software companies building AI apps and agents via App Accelerate: App Accelerate supports software companies building AI apps and agents on the Microsoft agent stack, with benefits designed to bring agentic solutions to market with strong foundations for trust.
Investments in skilling
This year we are investing in partner skilling that connects certification readiness to project-ready execution, with role-based experiences like Project Ready Workshops that translate skills into repeatable delivery practices. These learning experiences are delivered through our Partner Skilling Hub.
We are also introducing the Frontier Engineer Badge, a new learning path delivered through Titan Academy that prepares Solution Engineers and Solution Architects within the partners’ organization to design, build and operate production-ready agentic AI solutions across the Frontier Transformation stack, including Microsoft Copilot, Copilot Studio, Azure AI Foundry, GitHub Copilot, Microsoft Fabric and Agent 365.
The journey is hands-on by design and follows a three-part model: earn required certifications to establish a shared technical baseline, demonstrate delivery capability through Project Ready (building and integrating agents with governance, security and compliance) and build advanced readiness for operating at scale through governance, velocity and industry solution patterns. The outcome is clear: delivery-ready engineers who can move customers from prototypes to trusted, governed deployments.
Capturing the Marketplace opportunity
Marketplaces matter more in 2026 as customers consolidate procurement and expect faster time to value, especially as AI moves from pilots to production. With over 5,000 AI solutions available, Microsoft Marketplace increases discoverability for partner-built AI solutions, including agents, and supports a more repeatable buy-and-deploy motion through familiar procurement. It also makes it easy for partners to package multiparty software and services offers, so customers can purchase what they need to implement, govern and scale AI in production.
Omdia projects Microsoft Marketplace as a nearly $300 billion partner services opportunity by 2030. In the same study, partners selling through Marketplace reported go-to-market benefits, including faster sales cycles and larger deals, with 75% of study participants reporting faster closes and 69% reporting larger deals through Microsoft Marketplace.(5)
To accelerate demand generation and make it easy to activate these motions, we recently introduced Partner Marketing Center Pro, an AI-powered experience for end-to-end campaign creation. It brings campaign discovery, customization and co-branding, intelligent localization and translation, automated publishing and built-in reporting into one workflow, with an AI assistant that provides coaching throughout the process.
Partner Marketing Center Pro is a benefit of the Microsoft AI Cloud Partner Program available to partners who have purchased at least one partner benefits package, who have attained a Solutions Partner designation or who are currently enrolled in ISV Success.
Ways to engage now
Here are a few practical next steps partners can take to maximize their Microsoft investment:
Start by joining the Microsoft AI Cloud Partner Program.
- Plan to attend our annual sales kickoff for partners, virtually on Wednesday, July 22, 2026, to learn more about FY27 priorities and execution motions.
- Explore partner benefits packages and select the option that aligns to their growth plan.
- Pursue a Solutions Partner designation or specialization that matches their solution plays and customer demand.
- Visit the Partner Skilling Hub to find role-based learning resources.
- Launch an FY26 campaign focused on SMB, security, Copilot adoption or marketplace growth using Partner Marketing Center Pro.
Making Frontier Transformation real
Frontier Transformation is about building AI-powered operating capability grounded in intelligence and trust, and delivered consistently across industries, geographies and market segments. Partners make that real for customers by turning strategy into production-ready solutions, with governance, security and adoption built in from day one.
Microsoft is committed to partner success. We will continue investing in the Microsoft AI Cloud Partner Program, with the incentives, the skilling and the go-to-market capabilities that enable partners to build repeatable offers, increase discoverability and deliver trusted AI outcomes for customers at scale.
Nicole Dezen leads the Microsoft partner ecosystem and the Global Channel Partner Sales organization in Small, Medium Enterprises and Channel (SME&C). As Chief Partner Officer, she has grown the Microsoft partner ecosystem to become the largest in the industry, enabling more than 500,000 partners to deliver AI transformation to millions of customers in each segment around the world.
Footnotes
1 Microsoft FY25 Third Quarter Earnings Conference Call, Microsoft, April 2025.
2 IDC Info Snapshot, sponsored by Microsoft, 1.3 Billion AI Agents by 2028, #US53361825, May 2025.
3 Based on Microsoft first party telemetry measuring agents built with Microsoft Copilot Studio or Microsoft Agent Builder that were in use during the last 28 days of November 2025.
4 Omdia, Unlocking Growth Potential: Partner Perspectives on Microsoft CSP, December 2025. Results are not an endorsement of Microsoft. Any reliance on these results is at the third party’s own risk.
5 Microsoft estimate based on IDC data (SMB TAM: $777B by FY26; $1T+ by 2030), as published on the Microsoft Partner Blog, “The Microsoft Marketplace opportunity for channel ecosystem,” November 20, 2025.
6 Omdia, Partner Ecosystem Multiplier – The Microsoft Marketplace Opportunity, commissioned research sponsored by Microsoft, December 2025. Results are not an endorsement of Microsoft. Any reliance on these results is at the third party’s own risk.
Throughout this document, $ refers to USD.
The post Accelerating Frontier Transformation with Microsoft partners appeared first on The Official Microsoft Blog.
AI has moved quickly from experimentation to production. Customers want measurable business outcomes, along with security, governance and responsible AI built in from day one. Microsoft partners are a meaningful differentiator to deliver these objectives. They turn ideas into deployable solutions by prioritizing the highest value use cases, building the right data and security foundations…
The post Accelerating Frontier Transformation with Microsoft partners appeared first on The Official Microsoft Blog.Read More
Why are Name‑Value arguments not visible at function entry breakpoints in MATLAB R2025a?
I am using argument validation in a function that has multiple inputs including Name-Value pair arguments. When I set a breakpoint at the beginning of this function, only the first argument appears in the MATLAB R2025a Workspace immediately. The rest do not appear until after the arguments block finishes running. However, if I remove the Name-Value arguments, all of the values appear immediately.
Why does this occur? Is there a workaround?I am using argument validation in a function that has multiple inputs including Name-Value pair arguments. When I set a breakpoint at the beginning of this function, only the first argument appears in the MATLAB R2025a Workspace immediately. The rest do not appear until after the arguments block finishes running. However, if I remove the Name-Value arguments, all of the values appear immediately.
Why does this occur? Is there a workaround? I am using argument validation in a function that has multiple inputs including Name-Value pair arguments. When I set a breakpoint at the beginning of this function, only the first argument appears in the MATLAB R2025a Workspace immediately. The rest do not appear until after the arguments block finishes running. However, if I remove the Name-Value arguments, all of the values appear immediately.
Why does this occur? Is there a workaround? namevalue, arguments, validation, breakpoint, workspace MATLAB Answers — New Questions
How to add a header to a printed or published PDF from a MATLAB script in MATLAB R2025a?
I am printing a MATLAB script to PDF in MATLAB R2025a, but the header with the timestamp is missing. This header was included automatically in MATLAB R2024b. How can I add the header back in R2025a?I am printing a MATLAB script to PDF in MATLAB R2025a, but the header with the timestamp is missing. This header was included automatically in MATLAB R2024b. How can I add the header back in R2025a? I am printing a MATLAB script to PDF in MATLAB R2025a, but the header with the timestamp is missing. This header was included automatically in MATLAB R2024b. How can I add the header back in R2025a? header, print, matlab, editor, pdf, date, filename MATLAB Answers — New Questions
Error converting python DataFrame to Table
I have used the following commands to load in a python .pkl file.
fid = py.open("data.pkl");
data = py.pickle.load(fid);
T = table(data);
This loads a python DataFrame object. Newer versions of MATLAB have the ability to convert this object to a table using the table command, which I tried but encountered the below error:
Error using py.pandas.DataFrame/table
Dimensions of the key and value must be the same, or the value must be scalar.
What does this error mean? I’m guessing it’s because the DataFrame object in the .pkl contains a couple nested fields. Most of the fields are simply 1xN numeric vectors, but a couple are 1xN objects which then have their own fields.
How can I convert this DataFrame object to something usable in MATLAB? I was given this datafile and did not generate it, and I am much more proficient in MATLAB than python, so I would rather solve this within MATLAB rather than having to create a python script or change how the file is created.I have used the following commands to load in a python .pkl file.
fid = py.open("data.pkl");
data = py.pickle.load(fid);
T = table(data);
This loads a python DataFrame object. Newer versions of MATLAB have the ability to convert this object to a table using the table command, which I tried but encountered the below error:
Error using py.pandas.DataFrame/table
Dimensions of the key and value must be the same, or the value must be scalar.
What does this error mean? I’m guessing it’s because the DataFrame object in the .pkl contains a couple nested fields. Most of the fields are simply 1xN numeric vectors, but a couple are 1xN objects which then have their own fields.
How can I convert this DataFrame object to something usable in MATLAB? I was given this datafile and did not generate it, and I am much more proficient in MATLAB than python, so I would rather solve this within MATLAB rather than having to create a python script or change how the file is created. I have used the following commands to load in a python .pkl file.
fid = py.open("data.pkl");
data = py.pickle.load(fid);
T = table(data);
This loads a python DataFrame object. Newer versions of MATLAB have the ability to convert this object to a table using the table command, which I tried but encountered the below error:
Error using py.pandas.DataFrame/table
Dimensions of the key and value must be the same, or the value must be scalar.
What does this error mean? I’m guessing it’s because the DataFrame object in the .pkl contains a couple nested fields. Most of the fields are simply 1xN numeric vectors, but a couple are 1xN objects which then have their own fields.
How can I convert this DataFrame object to something usable in MATLAB? I was given this datafile and did not generate it, and I am much more proficient in MATLAB than python, so I would rather solve this within MATLAB rather than having to create a python script or change how the file is created. python, table, dataframe MATLAB Answers — New Questions
Microsoft 365 Quarterly Uptime Number Sinks to New Low
Microsoft 365 Uptime Suffers Major Blip in First Quarter of Calendar Year 2026
Microsoft’s service health and continuity page publishes details of Microsoft 365 Service Level Availability on a quarterly basis. The figures are worldwide for all commercial regions and reflects the combined uptime performance across all workloads. Microsoft doesn’t report details of uptime performance for individual regions. In effect, the figure is a metric for reporting and transparency purposes at a very high level instead of detailed information about a single tenant or even tenants in a single country.
Lowest Uptime Number Ever
The 99.526% figure published for Microsoft 365 uptime in the first quarter of the 2026 calendar year came as a real surprise because it was such a big drop over the performance delivered in the previous quarters (Figure 1).
Delivering 99.526% uptime means that service interruptions clocked up approximately 614 minutes (just over ten hours) in the quarter. Not only was the figure well below the norm (performance in Q4 CY2025 was 99.954%), but it also marked the lowest figure for uptime performance since the Office 365 for IT Pros team began to track the data in 2013 (Table 1).
| Q1 2013 | Q2 2013 | Q3 2013 | Q4 2013 | Q1 2014 | Q2 2014 | Q3 2014 | Q4 2014 |
| 99.94% | 99.97% | 99.96% | 99.98% | 99.99% | 99.95% | 99.98% | 99.99% |
| Q1 2015 | Q2 2015 | Q3 2015 | Q4 2015 | Q1 2016 | Q2 2016 | Q3 2016 | Q4 2016 |
| 99.99% | 99.95% | 99.98% | 99.98% | 99.98% | 99.98% | 99.99% | 99.99% |
| Q1 2017 | Q2 2017 | Q3 2017 | Q4 2017 | Q1 2018 | Q2 2018 | Q3 2018 | Q4 2018 |
| 99.99% | 99.97% | 99.985% | 99.988% | 99.993% | 99.98% | 99.97% | 99.98% |
| Q1 2019 | Q2 2019 | Q3 2019 | Q4 2019 | Q1 2020 | Q2 2020 | Q3 2020 | Q4 2020 |
| 99.97% | 99.97% | 99.98% | 99.98% | 99.98% | 99.99% | 99.97% | 99.97% |
| Q1 2021 | Q2 2021 | Q3 2021 | Q4 2021 | Q1 2022 | Q2 2022 | Q3 2022 | Q4 2022 |
| 99.97% | 99.98% | 99.985% | 99.976% | 99.98% | 99.98% | 99.99% | 99.99% |
| Q1 2023 | Q2 2023 | Q3 2023 | Q4 2023 | Q1 2024 | Q2 2024 | Q3 2024 | Q4 2024 |
| 99.98% | 99.99% | 99.99% | 99.996% | 99.97% | 99.99% | 99.977% | 99.927% |
| Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | Q1 2026 | |||
| 99.988% | 99.995% | 99.991% | 99.954% | 99.526% |
Microsoft’s published data for Microsoft 365 uptime only goes back to 2022, but we’ve noted the quarterly performance ever since Microsoft introduced its financially-backed guarantee for the Office 365 service (as it was at the time).
Impact on Uptime Doesn’t Mean Any Service Credits
A large drop in uptime performance in Q1 might seem staggering, but it is only an indicator that one or more largescale incidents occurred, perhaps in several regions. Certainly, the impact of the incidents was sufficient to materially affect the global uptime metric. It also doesn’t mean that Microsoft is on the hook to pay out service credits to its customers under the service level agreement for online services. When reading Microsoft documentation about availability, you might find this page to be a useful guide to how Microsoft thinks about Service Level Agreements.
Although Microsoft’s commitment is to deliver 99.9% uptime for its online services, the measurement on a per-service, per-month basis. Indeed, Calling Plans, Teams Phone, and Audio Conferencing go even higher with a 99.999% uptime, up from the 99.99% target set in 2021.
Customers can only claim service credits if a specific service like Exchange Online or SharePoint Online falls below the 99.9% threshold with the uptime percentage calculated as:
Delivering 99.526% uptime means that service interruptions clocked up approximately 614 minutes (just over ten hours) in the quarter. Not only was the figure well below the norm (performance in Q4 CY2025 was 99.954%), but it also marked the lowest figure for uptime performance since the Office 365 for IT Pros team began to track the data in 2013 (Table 1).
| Q1 2013 | Q2 2013 | Q3 2013 | Q4 2013 | Q1 2014 | Q2 2014 | Q3 2014 | Q4 2014 |
| 99.94% | 99.97% | 99.96% | 99.98% | 99.99% | 99.95% | 99.98% | 99.99% |
| Q1 2015 | Q2 2015 | Q3 2015 | Q4 2015 | Q1 2016 | Q2 2016 | Q3 2016 | Q4 2016 |
| 99.99% | 99.95% | 99.98% | 99.98% | 99.98% | 99.98% | 99.99% | 99.99% |
| Q1 2017 | Q2 2017 | Q3 2017 | Q4 2017 | Q1 2018 | Q2 2018 | Q3 2018 | Q4 2018 |
| 99.99% | 99.97% | 99.985% | 99.988% | 99.993% | 99.98% | 99.97% | 99.98% |
| Q1 2019 | Q2 2019 | Q3 2019 | Q4 2019 | Q1 2020 | Q2 2020 | Q3 2020 | Q4 2020 |
| 99.97% | 99.97% | 99.98% | 99.98% | 99.98% | 99.99% | 99.97% | 99.97% |
| Q1 2021 | Q2 2021 | Q3 2021 | Q4 2021 | Q1 2022 | Q2 2022 | Q3 2022 | Q4 2022 |
| 99.97% | 99.98% | 99.985% | 99.976% | 99.98% | 99.98% | 99.99% | 99.99% |
| Q1 2023 | Q2 2023 | Q3 2023 | Q4 2023 | Q1 2024 | Q2 2024 | Q3 2024 | Q4 2024 |
| 99.98% | 99.99% | 99.99% | 99.996% | 99.97% | 99.99% | 99.977% | 99.927% |
| Q1 2025 | Q2 2025 | Q3 2025 | Q4 2025 | Q1 2026 | |||
| 99.988% | 99.995% | 99.991% | 99.954% | 99.526% |
Table 1: Microsoft 365 SLA performance since 2013 (from Office 365 for IT Pros)
Microsoft’s online data only goes back to 2022, but we’ve noted the quarterly performance ever since Microsoft introduced its financially-backed guarantee for the Office 365 service (as it was at the time).
Impact on Uptime Doesn’t Mean Any Service Credits
A large drop in uptime performance in Q1 might seem staggering, but it is only an indicator that one or more largescale incidents occurred, perhaps in several regions. Certainly, the impact of the incidents was sufficient to materially affect the global uptime metric. It also doesn’t mean that Microsoft is on the hook to pay out service credits to its customers under the service level agreement for online services. When reading Microsoft documentation about availability, you might find this page to be a useful guide to how Microsoft thinks about Service Level Agreements.
Although Microsoft’s commitment is to deliver 99.9% uptime for its online services, the measurement on a per-service, per-month basis. Indeed, Calling Plans, Teams Phone, and Audio Conferencing go even higher with a 99.999% uptime, up from the 99.99% target set in 2021.
Customers can only claim service credits if a specific service like Exchange Online or SharePoint Online falls below the 99.9% threshold with the uptime percentage calculated as:
Claims cannot be made based on quarterly performance nor across multiple workloads. Given the work involved to gather evidence and logs, many tenants don’t bother pursuing claims, and to be fair to Microsoft, the uptime performance for online services is usually very good.
Spread of Microsoft 365 and Large Number of Users Makes Uptime Hard to Imagine
According to Microsoft’s FY26 Q2 results, Microsoft 365 now “exceeds 450 million paid seats” Taking the uptime calculation for a service, each user had 129,600 minutes to access online services in Q1 CY26. Microsoft 365 as a whole has 58.32 trillion user minutes. Losing 0.00474% of those minutes means that Microsoft 365 tenants as a whole lost approximately 276 billion user minutes in the quarter (by comparison, the figure for Q4 CY25 is approximately 27 million).
Microsoft 365 service interruptions happen all the time. The question is whether an outage affects your tenant. Given the distributed nature of the Microsoft 365 datacenter network, high availability features incorporated into workloads (like Exchange Database Availability Groups), and the distribution of tenants and workloads across the network, a single outage seldom affects more than one region. The blip in uptime availability in Q1 CY26 is a blip, but no more than that. At least, that’s what the historic uptime data tells us. Time will tell.
Support the work of the Office 365 for IT Pros team by subscribing to the Office 365 for IT Pros eBook. Your support pays for the time we need to track, analyze, and document the changing world of Microsoft 365 and Office 365. Only humans contribute to our work!
How to put a title on a colorbar?
I have a 3D surface surf(X,Y,Z) viewed from view(0,90) with a colorbar which I want to put a title on. The help instructions talk about an lcolorbar, TitleString and ZlabelString but there’s no example and I’m lost.
[X Y]=meshgrid(0:100,0:100);
Z=Y;
surf(X,Y,Z);
view(0,90);
hcb=colorbar;
?????? what next to put a title on the colorbar please ?????
Maybe something like set(get(hcb,’Title’),’cb title’) but I wouldn’t be asking if that worked …
Thanks.I have a 3D surface surf(X,Y,Z) viewed from view(0,90) with a colorbar which I want to put a title on. The help instructions talk about an lcolorbar, TitleString and ZlabelString but there’s no example and I’m lost.
[X Y]=meshgrid(0:100,0:100);
Z=Y;
surf(X,Y,Z);
view(0,90);
hcb=colorbar;
?????? what next to put a title on the colorbar please ?????
Maybe something like set(get(hcb,’Title’),’cb title’) but I wouldn’t be asking if that worked …
Thanks. I have a 3D surface surf(X,Y,Z) viewed from view(0,90) with a colorbar which I want to put a title on. The help instructions talk about an lcolorbar, TitleString and ZlabelString but there’s no example and I’m lost.
[X Y]=meshgrid(0:100,0:100);
Z=Y;
surf(X,Y,Z);
view(0,90);
hcb=colorbar;
?????? what next to put a title on the colorbar please ?????
Maybe something like set(get(hcb,’Title’),’cb title’) but I wouldn’t be asking if that worked …
Thanks. colorbar title surf titlestring MATLAB Answers — New Questions
UAV Toolbox Support Package for PX4 Autopilots — SITL plant simulation running much slower after recent windows update
I’ve been developing a custom controller using this toolbox, and everything has been working well after following the examples provided in the documentation.
However, after this last Tuesday, April 14th 2026, I attempted to monitor and tune my controller, and it threw an error saying the build failed and I need to go through the toolbox setup again (this is the first time I’ve gotten this error). I did this, and that is when problems developed. When using a simulink plant simulation for the PX4 SITL Host, the plant can no longer run in real time. It runs at about a quarter of the speed. I booted up the PX4 monitor and tune with plant example, and even that example is now running slowly too! Is it possible the recent security update affected TCP communication between Windows and the WSL network adapter?
Also, I am on R2025a.I’ve been developing a custom controller using this toolbox, and everything has been working well after following the examples provided in the documentation.
However, after this last Tuesday, April 14th 2026, I attempted to monitor and tune my controller, and it threw an error saying the build failed and I need to go through the toolbox setup again (this is the first time I’ve gotten this error). I did this, and that is when problems developed. When using a simulink plant simulation for the PX4 SITL Host, the plant can no longer run in real time. It runs at about a quarter of the speed. I booted up the PX4 monitor and tune with plant example, and even that example is now running slowly too! Is it possible the recent security update affected TCP communication between Windows and the WSL network adapter?
Also, I am on R2025a. I’ve been developing a custom controller using this toolbox, and everything has been working well after following the examples provided in the documentation.
However, after this last Tuesday, April 14th 2026, I attempted to monitor and tune my controller, and it threw an error saying the build failed and I need to go through the toolbox setup again (this is the first time I’ve gotten this error). I did this, and that is when problems developed. When using a simulink plant simulation for the PX4 SITL Host, the plant can no longer run in real time. It runs at about a quarter of the speed. I booted up the PX4 monitor and tune with plant example, and even that example is now running slowly too! Is it possible the recent security update affected TCP communication between Windows and the WSL network adapter?
Also, I am on R2025a. px4, uav toolbox support package, monitor and tune, simulink, plant MATLAB Answers — New Questions
I want to install Matlab on my desktop with connection from Network license server . Steps required
I want to install Matlab on my desktop with connection from Network license server . Steps requiredI want to install Matlab on my desktop with connection from Network license server . Steps required I want to install Matlab on my desktop with connection from Network license server . Steps required client installation MATLAB Answers — New Questions
Evaluate Noise and vibration of an EV motor (PMSM)
I want to find out the Noise and vibration of an EV motor PMSM by given few input data and getting results. So what is the process in the matlab/simulink to get the resultsI want to find out the Noise and vibration of an EV motor PMSM by given few input data and getting results. So what is the process in the matlab/simulink to get the results I want to find out the Noise and vibration of an EV motor PMSM by given few input data and getting results. So what is the process in the matlab/simulink to get the results nvh of an ev motor MATLAB Answers — New Questions
The Microsoft Graph PowerShell SDK and the additionalProperties Property
What is the additionalProperties Property and What is Its Purpose?
Experienced PowerShell developers who are accustomed to working with modules like the Exchange Online management module or old AzureAD module often express surprise when they see how Microsoft Graph PowerShell SDK cmdlets return objects. Anyone who’s used to getting objects full of nice well-formed properties might be confused when they discover that cmdlets like Get-MgUserMemberOf and Get-MgGroupMember only return a couple of properties, including the identifiers (GUIDs) for matching objects. The immediate conclusion might be that the identifiers are necessary to retrieve the full set of object properties with a separate call. In fact, the cmdlets usually return additional information for objects in a property aptly named additionalProperties.
This behavior naturally creates the question why does the Microsoft Graph PowerShell SDK use the additionalProperties property to hold so much valuable information? The reason lies in the way that the AutoRest process generates Graph SDK cmdlets from the OpenAPI descriptions for Graph APIs.
AutoRest and OpenAPI Descriptions
Microsoft uses a process called AutoRest to convert the OpenAPI specifications for Graph APIs into the modules and cmdlets that form the Microsoft Graph PowerShell SDK (AutoRest is also used to create other SDKs).
When AutoRest runs to create a new version of the Microsoft Graph PowerShell SDK, it checks the output that each cmdlet should produce. However, many of the output properties for Graph queries are dynamic and cannot be converted by AutoRest into strongly-typed PowerShell properties. AutoRest works around this problem by making cmdlets generate normal PowerShell properties for Graph properties of a known type and capturing anything that isn’t a known typed property (like an array or string) in a catch-all PowerShell dictionary called additionalProperties. This implementation is in line with the OpenAPI rules for handling data that is not explicitly defined in the schema.
For example, the Get-MgDirectoryDeletedItem cmdlet fetches details of soft-deleted Entra ID objects. The only strongly-typed properties defined for the cmdlet are Id and DeletedDateTime, which is why we see these properties when the cmdlet is run to fetch details of a soft-deleted group:
Get-MgDirectoryDeletedItem -DirectoryObjectId $GroupId Id DeletedDateTime -- --------------- bfa786a7-2394-495b-9e9e-a8b52c9ee3be 07/03/2026 12:38:30
Most of the properties describing the deleted group is in additionalProperties:
(Get-MgDirectoryDeletedItem -DirectoryObjectId $GroupId).additionalProperties
Key Value
--- -----
@odata.context https://graph.microsoft.com/v1.0/$metadata#directoryObjects/$entity
@odata.type #microsoft.graph.group
createdDateTime 2021-11-01T17:52:57Z
creationOptions {}
description Case studies
displayName Customer Case Studies
expirationDateTime 2027-08-22T08:07:44Z
However, if you run the Get-MgDirectoryDeletedGroup cmdlet instead of Get-MgDirectoryDeletedItem, more properties are returned. As discussed below, this is because Get-MgDirectoryDeletedItem can process many different types of soft-deleted Entra ID objects, so the set of properties it outputs are those common to all objects. By comparison, Get-MgDirectoryDeletedGroup only handles soft-deleted groups, so the set of strongly-typed properties is larger:
Get-MgDirectoryDeletedGroup | Format-Table DisplayName, Description, Id DisplayName Description Id ----------- ----------- -- Azure AD Testers People who do early testing of Azure AD features 36fcfd60-9ad8-48ed-8c3c-ef36fc5d0c94
The additionalProperties Hash Table
A PowerShell dictionary is a hash table composed of key value pairs, so the value of a property in the dictionary can be accessed by referencing its name (the key). Be careful with casing. Sometimes it matters and sometimes it doesn’t. It’s safest to use the casing used by the cmdlet when it outputs values held in additionalProperties:
(Get-MgDirectoryDeletedItem -DirectoryObjectId $GroupId).additionalProperties.displayName Customer Case Studies
Similar output is seen in many other Graph SDK cmdlets. For example, here’s the output from Get-MgGroupMember:
Get-MgGroupMember -GroupId $GroupId Id DeletedDateTime -- --------------- 5b52fba5-349e-4624-88cd-d790883fe4c4 a221d10f-e0cf-4a1d-b6a2-4e844670f118 7bfd3f83-be63-4a5a-bbf8-c821e2836920 70ba4f9f-c357-4f08-a746-5d4d03794e3d
To see the display names for the group members, fetch the information from additionalProperties:
(Get-MgGroupMember -GroupId $GroupId).additionalProperties.displayName Ken Bowers Chris Bishop Alain Charnier Marty King
A common requirement for scripts is to create a comma-separated string from properties for report output. The details returned for a group by the Get-MgGroup cmdlet don’t include its owners. To fetch the display names of the group owners, run the Get-MgGroupOwner cmdlet like this:
$Group = (Get-MgGroup -Filter "displayName eq 'Ultra Fans'")
[array]$Owners = Get-MgGroupOwner -GroupId $Group.Id | Select-Object
-ExpandProperty additionalProperties
$OwnersOutput = $Owners.displayName -join ", "
Write-Host ("The owners of the {0} group are {1}" -f $Group.displayName, $OwnersOutput)
You’ll find many examples of using the information held in the additionalProperties hash table in scripts in the Office 365 for IT Pros GitHub repository. For example, the script to replace group owners with PowerShell uses the property in conjunction with the Get-MgUserOwnedObject cmdlet to determine the set of Microsoft 365 groups owned by an account.
When Microsoft Graph PowerShell SDK Cmdlets Generate the additionalProperties Property
AdditionalProperties are encountered most often when:
- A cmdlet can return multiple object types: For instance, the membership of an administrative unit can include user accounts, devices, and groups. The different object types can be distinguished in additionalProperties by the value of the odata.type key. This is the situation when the Get-MgDirectoryDeletedItem cmdlet is used as described above.
- A cmdlet is requested to output a property that is not common to all the object types retrieved by a cmdlet. For example, the City property is available for user objects, but not for groups, so if you include City in the set of properties retrieved by Get-MgGroupMember, the output for the City property is in additionalProperties (but only for user objects).
- A cmdlet expands a navigation property. For example, a user account has a Manager property which points (navigates) to the user account for the user’s Manager. By default, the Manager property is not in the set fetched by the Get-MgUser cmdlet. However, if the Get-MgUser cmdlet includes the ExpandProperty parameter to expand the Manager property, the cmdlet outputs the Manager property with additionalProperties containing details of the manager’s account (Figure 1).
Solving an AutoRest Problem
In summary, the additionalProperties property solves a problem for AutoRest with Graph properties which AutoRest can’t figure out how to deal with. The solution works, even if stuffing the properties and their values into the additionalProperties hash table seems a little strange when first encountered. It’s another of the little foibles that make the Microsoft Graph PowerShell SDK so “interesting” to work with.
Need help to write and manage PowerShell scripts for Microsoft 365, including Azure Automation runbooks? Get a copy of the Automating Microsoft 365 with PowerShell eBook, available standalone or as part of the Office 365 for IT Pros eBook bundle.
Stateflow transition does not work when using output bus field
Hello together,
I am facing a strange behaviour of my Simulink model. That model is comprised of a Simscape subsystem and a Stateflow subsystem. The latter is used to controll the operation mode of the Simscape model.
For a clear data transfer I implemented several data buses as Stateflow inputs and outputs. For that, I created the necessary Simulink Bus Objects and it works fine so far. But when I want to use one field of the Getoperationdata bus to describe the transition condition, the transition is not exceeded despite the condition is true (as shown in the figure below). The state Empty_Pump_Chamber remains active but should switch to Empty_Complete.
The Getoperationdata bus is implemented as an output bus. When defining a transition condition with a variable from another input bus, the model succeeds. Very strange. Even when I define a local variable, set this variable equal to Getoperationaldata.emptyComplete2 and use this new variable to state the transition condition, the model succeeds. Very strange! What is wrong?
Kind Regards!Hello together,
I am facing a strange behaviour of my Simulink model. That model is comprised of a Simscape subsystem and a Stateflow subsystem. The latter is used to controll the operation mode of the Simscape model.
For a clear data transfer I implemented several data buses as Stateflow inputs and outputs. For that, I created the necessary Simulink Bus Objects and it works fine so far. But when I want to use one field of the Getoperationdata bus to describe the transition condition, the transition is not exceeded despite the condition is true (as shown in the figure below). The state Empty_Pump_Chamber remains active but should switch to Empty_Complete.
The Getoperationdata bus is implemented as an output bus. When defining a transition condition with a variable from another input bus, the model succeeds. Very strange. Even when I define a local variable, set this variable equal to Getoperationaldata.emptyComplete2 and use this new variable to state the transition condition, the model succeeds. Very strange! What is wrong?
Kind Regards! Hello together,
I am facing a strange behaviour of my Simulink model. That model is comprised of a Simscape subsystem and a Stateflow subsystem. The latter is used to controll the operation mode of the Simscape model.
For a clear data transfer I implemented several data buses as Stateflow inputs and outputs. For that, I created the necessary Simulink Bus Objects and it works fine so far. But when I want to use one field of the Getoperationdata bus to describe the transition condition, the transition is not exceeded despite the condition is true (as shown in the figure below). The state Empty_Pump_Chamber remains active but should switch to Empty_Complete.
The Getoperationdata bus is implemented as an output bus. When defining a transition condition with a variable from another input bus, the model succeeds. Very strange. Even when I define a local variable, set this variable equal to Getoperationaldata.emptyComplete2 and use this new variable to state the transition condition, the model succeeds. Very strange! What is wrong?
Kind Regards! simulink bus object, transition MATLAB Answers — New Questions
Only version 2026a appears available for download as a trial. I want version 2025b
Only version 2026a appears available for download as a trial.
I want version 2025b, but it does not appear as an option at https://www.mathworks.com/downloads/web_downloads/13799194.Only version 2026a appears available for download as a trial.
I want version 2025b, but it does not appear as an option at https://www.mathworks.com/downloads/web_downloads/13799194. Only version 2026a appears available for download as a trial.
I want version 2025b, but it does not appear as an option at https://www.mathworks.com/downloads/web_downloads/13799194. r2025a MATLAB Answers — New Questions
Why the figure is not apearing after using plot(x) command?
Hi,
after new instalation of MATLAB 2025b my program does not use function plot properly.
After command "plot(x)" MATLAB opens new window "Figures" as usual but there is no figure.
After this line program continues execution.
I could not find the explanation jet.
Thanks!Hi,
after new instalation of MATLAB 2025b my program does not use function plot properly.
After command "plot(x)" MATLAB opens new window "Figures" as usual but there is no figure.
After this line program continues execution.
I could not find the explanation jet.
Thanks! Hi,
after new instalation of MATLAB 2025b my program does not use function plot properly.
After command "plot(x)" MATLAB opens new window "Figures" as usual but there is no figure.
After this line program continues execution.
I could not find the explanation jet.
Thanks! plot, figures MATLAB Answers — New Questions
I can’t able to see the orange, gray, red or green checks. Instead I can see only the code metrics.
I’m facing the issue i.e., I’m running the polyspace code prover throught the commend prompt. I used all the configuration in the option.txt file. Through that option.txt file i’m start compling by using this cmd ("C:Program FilesPOLYSPACER2023apolyspacebinpolyspace-code-prover" ^ -options-file Options.txt ^ -scheduler CHE6-SV00106 ^ -results-dir "C:UsersaallimutDocumentsSWEET500cb1146_vcore_main_devResult"). Compliation is done sucessfully and the code is pushed to the server. After successful complition of run I Checked the result, It’s only showing the code metrix, not showing any checks in the dashboard.I’m facing the issue i.e., I’m running the polyspace code prover throught the commend prompt. I used all the configuration in the option.txt file. Through that option.txt file i’m start compling by using this cmd ("C:Program FilesPOLYSPACER2023apolyspacebinpolyspace-code-prover" ^ -options-file Options.txt ^ -scheduler CHE6-SV00106 ^ -results-dir "C:UsersaallimutDocumentsSWEET500cb1146_vcore_main_devResult"). Compliation is done sucessfully and the code is pushed to the server. After successful complition of run I Checked the result, It’s only showing the code metrix, not showing any checks in the dashboard. I’m facing the issue i.e., I’m running the polyspace code prover throught the commend prompt. I used all the configuration in the option.txt file. Through that option.txt file i’m start compling by using this cmd ("C:Program FilesPOLYSPACER2023apolyspacebinpolyspace-code-prover" ^ -options-file Options.txt ^ -scheduler CHE6-SV00106 ^ -results-dir "C:UsersaallimutDocumentsSWEET500cb1146_vcore_main_devResult"). Compliation is done sucessfully and the code is pushed to the server. After successful complition of run I Checked the result, It’s only showing the code metrix, not showing any checks in the dashboard. no checks found MATLAB Answers — New Questions
Identifying bright spots in RGB image
I have hundreds of RGB images (one of the image is attached). For reference these are interference images and the RGB color values correspond to certain film thickness between a semi reflective glass surface and a steel surface. I apply electric potential across the surface and observed electric discharges which show as bright spots on the image (highlighted in red circle in attached image).
I want to write a script in MATLAB (2023a) that goes through the images, identify these bright spots stores there location in the XY cordinate system. The end goal is to show spatial distribution of the electric discharges.I have hundreds of RGB images (one of the image is attached). For reference these are interference images and the RGB color values correspond to certain film thickness between a semi reflective glass surface and a steel surface. I apply electric potential across the surface and observed electric discharges which show as bright spots on the image (highlighted in red circle in attached image).
I want to write a script in MATLAB (2023a) that goes through the images, identify these bright spots stores there location in the XY cordinate system. The end goal is to show spatial distribution of the electric discharges. I have hundreds of RGB images (one of the image is attached). For reference these are interference images and the RGB color values correspond to certain film thickness between a semi reflective glass surface and a steel surface. I apply electric potential across the surface and observed electric discharges which show as bright spots on the image (highlighted in red circle in attached image).
I want to write a script in MATLAB (2023a) that goes through the images, identify these bright spots stores there location in the XY cordinate system. The end goal is to show spatial distribution of the electric discharges. image analysis, image processing MATLAB Answers — New Questions
