Category Archives: Microsoft Blog

Microsoft Blog

Announcing Windows 10 Insider Preview Build 18362.10000 (19H2) | Windows Experience Blog

Hello Windows Insiders!Today, we are talking about the next steps we are taking at continuing to evolve Windows 10 servicing and quality with the next feature update for Windows 10. This release has been referred to as 19H2 with Insiders. 19H2 will include a scoped set of features for select performance improvements, enterprise features, and quality enhancements and will be delivered to customers running the May 2019 Update using servicing technology (like the monthly Cumulative Update process).
For Insiders, this means a few differences to how they will get 19H2:
Insiders will need to be in the Slow ring and on the May 2019 Update to receive 19H2 bits.
19H2 will be delivered to Insiders in the Slow ring via servicing as a Cumulative Update and not full build updates.
Some Insiders may not see the new features right away as we perform controlled feature rollouts (CFRs) to gain better feedback on overall build quality.
Specific to CFRs, we may ship features in these updates turned off by default and turn them on independently of bits getting downloaded to Insiders’ PCs.
Even though we will be delivering 19H2 to Insiders in the Slow ring through servicing, we plan to continue to publish blog posts for each of these flights and include information on anything new and notable and any known issues that might be included. For more information on what 19H2 means for Windows 10 servicing and quality, we recommend you read John Cable’s blog post here.
We are also releasing 19H2 Build 18362.10000 to Insiders in the Slow ring today. This update contains two behind-the-scenes changes designed for OEMs and does not contain anything visible to Insiders. We’re using this update to test our process and servicing pipeline for delivering these updates to customers. 19H2 updates to Insiders will also be cumulative with the latest May 2019 Updates. For example, today’s release also includes the same improvements and fixes contained in this Cumulative Update released for the May 2019 Update here. Going forward, these updates will continue to contain the same improvements and fixes released for the May 2019 Update in addition to new 19H2 changes.
As always, Insiders are encouraged to report any issues they experience with these updates through Feedback Hub.
Thanks,Dona and Brandon

Evolving Windows 10 servicing and quality: the next steps | Windows Experience Blog

Today, as part of our commitment to transparency, we are providing an overview of how we plan to further optimize the delivery of our next feature update. This optimization is specific to devices running the Windows 10 May 2019 Update. For devices running earlier versions of Windows 10, the process remains unchanged.In April, we announced enhancements to the Windows 10 May 2019 Update experience with an increased focus on user control, quality and transparency. These were a collection of improvements based on requests from customers like you to make the Windows update experience better with less disruption. With the next Windows 10 feature update, we’re taking this further.
Next feature release of Windows 10
The next feature update for Windows 10 (known in the Windows Insider Program as 19H2) will be a scoped set of features for select performance improvements, enterprise features and quality enhancements. To deliver these updates in a less disruptive fashion, we will deliver this feature update in a new way, using servicing technology (like the monthly update process) for customers running the May 2019 Update who choose to update to the new release. In other words, anyone running the May 2019 Update and updating to the new release will have a far faster update experience because the update will install like a monthly update.
For consumer or commercial users coming from versions of Windows 10 earlier than the May 2019 Update (version 1903), the process of updating to the new release will be the same as it has been and work in a similar manner to previous Windows 10 feature updates, using the same tools and processes.
As this release is a September-targeted release of Windows, commercial customers using Windows 10 Enterprise and Education editions of version 19H2 will continue to enjoy 30 months of servicing; for more specific information see this blog post.
Next steps
As with all our feature updates, we utilize a multifaceted quality strategy that includes automated and manual testing and leverages the Windows Insider Program to obtain user feedback and data on quality. The next update to Windows 10 will be no different. We will begin releasing 19H2 builds to Windows Insiders in the Slow ring starting today, with new features being offered in future Insider builds as they are ready. Note: some Insiders may not see the new features right away as we are using a controlled feature rollout (CFR)1 to gain better feedback on overall build quality. Broad availability of the next update to Windows 10 will begin later this calendar year. We will share further details on Insider Preview builds as we release each new build for both 19H2 (Slow ring) and 20H1 (Fast ring). Windows Insiders who have opted into the Fast ring have been providing feedback on 20H1 builds from our development branch since February 14.
We will provide more information on new features as we get closer to the launch of the next update to Windows 10. We are continuing to evolve how we deliver a great Windows 10 update experience to our customers and ecosystem and look forward to hearing your feedback.
Note:1 Controlled Feature Rollout (CFR): A method to progressively rollout new features by gradually increasing the audience in a controlled manner.

Windows 10 Tip: Your Phone app gives you more to do with messages and photos | Windows Experience Blog

Editor’s note: We’re back with the summer batch of weekly Windows 10 tips posts, which highlight some of the many helpful features that come with the Windows 10 May 2019 Update. We’ve been working hard behind the scenes to make your daily life easier with a streamlined update process, as well as clean and simple experiences for your desktop. 
Thanks to the Windows 10 May 2019 Update, you’ll see improvements to the Your Phone  app that debuted with the Windows 10 October 2018 Update.  
This is the app that gives you instant access to your recent Android phone’s photos and texts on your PC – no need to dig for your phone to text or email yourself photos. You can find out more about its original features in our post from earlier this year. 
There’s more you can do with your messages with the most recent update. You can now attach emojis, GIFs or images directly to your texts, as well as in-line reply to incoming messages.  
Check it out in action: 

You’re also able to send and receive deep link URLs in your messages and see unread messages and threads. 
In addition to conveniently dragging and dropping photos directly to an email, presentation or desktop, you’re now able to save images directly to your PC by right-clicking on the image. 
Find out more about Your Phone. 
If you like this kind of information, check out more Windows 10 Tips. 

TensorWatch: A debugging and visualization system for machine learning

TensorWatch

The rise of deep learning is accompanied by ever-increasing model complexity, larger datasets, and longer training times for models. When working on novel concepts, researchers often need to understand why training metrics are trending the way they are. So far, the available tools for machine learning training have focused on a “what you see is what you log” approach. As logging is relatively expensive, researchers and engineers tend to avoid it and rely on a few signals to guesstimate the cause of the patterns they see. At Microsoft Research, we’ve been asking important questions surrounding this very challenge: What if we could dramatically reduce the cost of getting more information about the state of the system? What if we had advanced tooling that could help researchers make more informed decisions effectively?

Introducing TensorWatch

We’re happy to introduce TensorWatch, an open-source system that implements several of these ideas and concepts. We like to think of TensorWatch as the Swiss Army knife of debugging tools with many advanced capabilities researchers and engineers will find helpful in their work. We presented TensorWatch at the 2019 ACM SIGCHI Symposium on Engineering Interactive Computing Systems.

Custom UIs and visualizations

The first thing you might notice when using TensorWatch is it extensively leverages Jupyter Notebook instead of prepackaged user interfaces, which are often difficult to customize. TensorWatch provides the interactive debugging of real-time training processes using either the composable UI in Jupyter Notebooks or the live shareable dashboards in Jupyter Lab. In addition, since TensorWatch is a Python library, you can also build your own custom UIs or use TensorWatch in the vast Python data science ecosystem. TensorWatch also supports several standard visualization types, including bar charts, histograms, and pie charts, as well as 3D variations.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

Streams, streams everywhere

One of the central premises of the TensorWatch architecture is we uniformly treat data and other objects as streams. This includes files, console, sockets, cloud storage, and even visualizations themselves. With a common interface, TensorWatch streams can listen to other streams, which enables the creation of custom data flow graphs. Using these concepts, TensorWatch trivially allows you to implement a variety of advanced scenarios. For example, you can render many streams into the same visualization, or one stream can be rendered in many visualizations simultaneously, or a stream can be persisted in many files, or not persisted at all. The possibilities are endless!

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

Lazy logging mode

With TensorWatch, we also introduce lazy logging mode. This mode doesn’t require explicit logging of all the information beforehand. Instead, you can have TensorWatch observe the variables. Since observing is basically free, you can track as many variables as you like, including large models or entire batches during the training. TensorWatch then allows you to perform interactive queries that run in the context of these variables and returns the streams as a result. These streams can then be visualized, saved, or processed as needed. For example, you can write a lambda expression that computes mean weight gradients in each layer in the model at the completion of each batch and send the result as a stream of tensors that can be plotted as a bar chart.

Phases of model development

At Microsoft Research, we care deeply about improving debugging capabilities in all phases of model development—pre-training, in-training, and post-training. Consequently, TensorWatch provides many features useful for pre- and post-training phases as well. We lean on several excellent open-source libraries to enable many of these features, which include model graph visualization, data exploration through dimensionality reduction, model statistics, and several prediction explainers for convolution networks.

Open source on GitHub

We hope TensorWatch helps spark further advances and ideas for efficiently debugging and visualizing machine learning and invite the ML community to participate in this journey via GitHub.

Go to Original Article
Author: Microsoft News Center

Quantum launches Distributed Cloud Services

Quantum Corp. has announced a new line of services and storage-as-a-software offerings called Distributed Cloud Services. According to Quantum, it was designed to enable an enterprise’s resources to focus on meeting business goals, rather than spending time managing storage.

Quantum’s new Cloud-Based Analytics software powers Distributed Cloud Services. Products are designed to send data about their environment using the Cloud-Based Analytics software, making them part of Distributed Cloud.

Using the Cloud-Based Analytics software, users or Quantum’s own support team can manage and monitor environments worldwide from one central location. Users can monitor their own environments or choose to have Quantum do it through Distributed Cloud Services.

According to Quantum, the driving force behind the creation of Distributed Cloud Services was the need for businesses to create, study and develop more, while having fewer IT and engineering resources, therefore looking to others to manage data storage infrastructure.

Quantum also announced Quantum Operational Services, which it claims provides cloudlike storage with on-premises control. Users manage daily storage operations with Quantum, with the hope of providing more reliability through monitoring and analysis.

Quantum claims the key benefits of using the Operational Services line are eliminating the weight that managing storage places on IT resources, reducing downtime to improve UX, keeping the control and security of an on-premises storage center and maximizing storage ROI.

Lastly, Quantum has added new storage as a service (SaaS) offerings to its portfolio. This is aimed at those who would prefer a pay-per-use subscription service for the Operational Services features. The benefit to this, according to Quantum, is that users only pay for the storage they use.

Other vendors that offer SaaS include Kaminario, which recently expanded its SaaS offerings with metered usage payments, disaster recovery (DR) and service usage on the public cloud.

SaaS is generally considered a good choice for small or midsize enterprises. It can save money by eliminating the need for personnel to implement and maintain a storage infrastructure, as well as reduce DR risks and provide long-term record retention.

All new offerings are now available.

Go to Original Article
Author:

For Sale – xeon 1220v5 CPU, xeon 2603v4 CPU, g1610t CPU, ECC 2133p 16GB RAM

Discussion in ‘Desktop Computer Classifieds‘ started by dirtypaws, Jun 20, 2019.

  1. dirtypaws

    dirtypaws

    Novice Member

    Joined:
    Sep 29, 2015
    Messages:
    38
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Sheffield
    Ratings:
    +2

    Xeon 1220V5 £90 (x2)
    Xeon 2603v4 £75
    Celeron G1610t £10
    16GB 2133p ECC RAM £80 (x2)
    [​IMG]

    Price and currency: £100
    Delivery: Delivery cost is not included
    Payment method: paypal , bank transfer
    Location: sheffield
    Advertised elsewhere?: Not advertised elsewhere
    Prefer goods collected?: I have no preference

    ______________________________________________________
    This message is automatically inserted in all classifieds forum threads.
    By replying to this thread you agree to abide by the trading rules detailed here.
    Please be advised, all buyers and sellers should satisfy themselves that the other party is genuine by providing the following via private conversation to each other after negotiations are complete and prior to dispatching goods and making payment:

    • Landline telephone number. Make a call to check out the area code and number are correct, too
    • Name and address including postcode
    • Valid e-mail address

    DO NOT proceed with a deal until you are completely satisfied with all details being correct. It’s in your best interest to check out these details yourself.

Share This Page

Loading…

Go to Original Article
Author:

TensorWatch: A debugging and visualization system for machine learning

TensorWatch

The rise of deep learning is accompanied by ever-increasing model complexity, larger datasets, and longer training times for models. When working on novel concepts, researchers often need to understand why training metrics are trending the way they are. So far, the available tools for machine learning training have focused on a “what you see is what you log” approach. As logging is relatively expensive, researchers and engineers tend to avoid it and rely on a few signals to guesstimate the cause of the patterns they see. At Microsoft Research, we’ve been asking important questions surrounding this very challenge: What if we could dramatically reduce the cost of getting more information about the state of the system? What if we had advanced tooling that could help researchers make more informed decisions effectively?

Introducing TensorWatch

We’re happy to introduce TensorWatch, an open-source system that implements several of these ideas and concepts. We like to think of TensorWatch as the Swiss Army knife of debugging tools with many advanced capabilities researchers and engineers will find helpful in their work. We presented TensorWatch at the 2019 ACM SIGCHI Symposium on Engineering Interactive Computing Systems.

Custom UIs and visualizations

The first thing you might notice when using TensorWatch is it extensively leverages Jupyter Notebook instead of prepackaged user interfaces, which are often difficult to customize. TensorWatch provides the interactive debugging of real-time training processes using either the composable UI in Jupyter Notebooks or the live shareable dashboards in Jupyter Lab. In addition, since TensorWatch is a Python library, you can also build your own custom UIs or use TensorWatch in the vast Python data science ecosystem. TensorWatch also supports several standard visualization types, including bar charts, histograms, and pie charts, as well as 3D variations.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

Streams, streams everywhere

One of the central premises of the TensorWatch architecture is we uniformly treat data and other objects as streams. This includes files, console, sockets, cloud storage, and even visualizations themselves. With a common interface, TensorWatch streams can listen to other streams, which enables the creation of custom data flow graphs. Using these concepts, TensorWatch trivially allows you to implement a variety of advanced scenarios. For example, you can render many streams into the same visualization, or one stream can be rendered in many visualizations simultaneously, or a stream can be persisted in many files, or not persisted at all. The possibilities are endless!

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

Lazy logging mode

With TensorWatch, we also introduce lazy logging mode. This mode doesn’t require explicit logging of all the information beforehand. Instead, you can have TensorWatch observe the variables. Since observing is basically free, you can track as many variables as you like, including large models or entire batches during the training. TensorWatch then allows you to perform interactive queries that run in the context of these variables and returns the streams as a result. These streams can then be visualized, saved, or processed as needed. For example, you can write a lambda expression that computes mean weight gradients in each layer in the model at the completion of each batch and send the result as a stream of tensors that can be plotted as a bar chart.

Phases of model development

At Microsoft Research, we care deeply about improving debugging capabilities in all phases of model development—pre-training, in-training, and post-training. Consequently, TensorWatch provides many features useful for pre- and post-training phases as well. We lean on several excellent open-source libraries to enable many of these features, which include model graph visualization, data exploration through dimensionality reduction, model statistics, and several prediction explainers for convolution networks.

Open source on GitHub

We hope TensorWatch helps spark further advances and ideas for efficiently debugging and visualizing machine learning and invite the ML community to participate in this journey via GitHub.

Go to Original Article
Author: Microsoft News Center

Pica8 PicOS upgrade enhances network security

Pica8 upgraded its Linux-based network operating system PicOS to include new capabilities to address network efficiency and security.

With this release, Pica8 PicOS interoperates with existing network access control (NAC) tools, enabling fully automated network access policy enforcement in an open networking deployment. According to Pica8, automation improves operational efficiency with a simplified UI and improved security posture.

According to Forrester, 46% of information employees use personal laptops and mobile devices for work, which creates an increasingly complex device landscape that must be managed with BYOD programs and increased network security.

The age of BYOD and IoT deployments create security challenges for enterprise networking, according to Pica8; as enterprises automate, simplify and modernize access networks, most customers have NAC systems in place. PicOS’ NAC integration support and a centralized policy-based access control for network access points these challenges, according to the vendor.

Building off its October 2018 support for the Dell EMC N3132PX-ON 2.5G/5G Mutigig PoE switch, Pica8 added PicOS availability for ISE and ClearPass, supporting the Open Network Install Environment standard that enables PicOS to deploy onto platforms. Additionally, PicOS is automatically included with the Packetfence 9.0 release.

Pica8’s PicOS open networking operating system is offered in an enterprise edition as well as an SDN edition. PicOS Enterprise Edition installs on 1G to 100G open switches and offers the most comprehensive support, including the Debian Linux distribution, Nymble, and Pica8’s CrossFlow capability. PicOS SDN Edition includes the Debian Linux distribution and Nymble, and uses OpenFlow 1.5’s User-Defined Fields for packet inspection.

Go to Original Article
Author:

For Sale – xeon 1220v5 CPU, xeon 2603v4 CPU, g1610t CPU, ECC 2133p 16GB RAM

Discussion in ‘Desktop Computer Classifieds‘ started by dirtypaws, Jun 20, 2019.

  1. dirtypaws

    dirtypaws

    Novice Member

    Joined:
    Sep 29, 2015
    Messages:
    38
    Products Owned:
    0
    Products Wanted:
    0
    Trophy Points:
    8
    Location:
    Sheffield
    Ratings:
    +2

    Xeon 1220V5 £90 (x2)
    Xeon 2603v4 £75
    Celeron G1610t £10
    16GB 2133p ECC RAM £80 (x2)
    [​IMG]

    Price and currency: £100
    Delivery: Delivery cost is not included
    Payment method: paypal , bank transfer
    Location: sheffield
    Advertised elsewhere?: Not advertised elsewhere
    Prefer goods collected?: I have no preference

    ______________________________________________________
    This message is automatically inserted in all classifieds forum threads.
    By replying to this thread you agree to abide by the trading rules detailed here.
    Please be advised, all buyers and sellers should satisfy themselves that the other party is genuine by providing the following via private conversation to each other after negotiations are complete and prior to dispatching goods and making payment:

    • Landline telephone number. Make a call to check out the area code and number are correct, too
    • Name and address including postcode
    • Valid e-mail address

    DO NOT proceed with a deal until you are completely satisfied with all details being correct. It’s in your best interest to check out these details yourself.

Share This Page

Loading…

Go to Original Article
Author:

TensorWatch: A debugging and visualization system for machine learning

TensorWatch

The rise of deep learning is accompanied by ever-increasing model complexity, larger datasets, and longer training times for models. When working on novel concepts, researchers often need to understand why training metrics are trending the way they are. So far, the available tools for machine learning training have focused on a “what you see is what you log” approach. As logging is relatively expensive, researchers and engineers tend to avoid it and rely on a few signals to guesstimate the cause of the patterns they see. At Microsoft Research, we’ve been asking important questions surrounding this very challenge: What if we could dramatically reduce the cost of getting more information about the state of the system? What if we had advanced tooling that could help researchers make more informed decisions effectively?

Introducing TensorWatch

We’re happy to introduce TensorWatch, an open-source system that implements several of these ideas and concepts. We like to think of TensorWatch as the Swiss Army knife of debugging tools with many advanced capabilities researchers and engineers will find helpful in their work. We presented TensorWatch at the 2019 ACM SIGCHI Symposium on Engineering Interactive Computing Systems.

Custom UIs and visualizations

The first thing you might notice when using TensorWatch is it extensively leverages Jupyter Notebook instead of prepackaged user interfaces, which are often difficult to customize. TensorWatch provides the interactive debugging of real-time training processes using either the composable UI in Jupyter Notebooks or the live shareable dashboards in Jupyter Lab. In addition, since TensorWatch is a Python library, you can also build your own custom UIs or use TensorWatch in the vast Python data science ecosystem. TensorWatch also supports several standard visualization types, including bar charts, histograms, and pie charts, as well as 3D variations.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

With TensorWatch—a debugging and visualization tool for machine learning—researchers and engineers can customize the user interface to accommodate a variety of scenarios. Above is an example of TensorWatch running in Jupyter Notebook, rendering a live chart from multiple streams produced by an ML training application.

Streams, streams everywhere

One of the central premises of the TensorWatch architecture is we uniformly treat data and other objects as streams. This includes files, console, sockets, cloud storage, and even visualizations themselves. With a common interface, TensorWatch streams can listen to other streams, which enables the creation of custom data flow graphs. Using these concepts, TensorWatch trivially allows you to implement a variety of advanced scenarios. For example, you can render many streams into the same visualization, or one stream can be rendered in many visualizations simultaneously, or a stream can be persisted in many files, or not persisted at all. The possibilities are endless!

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

TensorWatch supports a variety of visualization types. Above is an example of a TensorWatch t-SNE visualization of the MNIST dataset.

Lazy logging mode

With TensorWatch, we also introduce lazy logging mode. This mode doesn’t require explicit logging of all the information beforehand. Instead, you can have TensorWatch observe the variables. Since observing is basically free, you can track as many variables as you like, including large models or entire batches during the training. TensorWatch then allows you to perform interactive queries that run in the context of these variables and returns the streams as a result. These streams can then be visualized, saved, or processed as needed. For example, you can write a lambda expression that computes mean weight gradients in each layer in the model at the completion of each batch and send the result as a stream of tensors that can be plotted as a bar chart.

Phases of model development

At Microsoft Research, we care deeply about improving debugging capabilities in all phases of model development—pre-training, in-training, and post-training. Consequently, TensorWatch provides many features useful for pre- and post-training phases as well. We lean on several excellent open-source libraries to enable many of these features, which include model graph visualization, data exploration through dimensionality reduction, model statistics, and several prediction explainers for convolution networks.

Open source on GitHub

We hope TensorWatch helps spark further advances and ideas for efficiently debugging and visualizing machine learning and invite the ML community to participate in this journey via GitHub.

Go to Original Article
Author: Microsoft News Center