News For Dev-ops

Starting today, the Amazon Elastic Compute Cloud (Amazon EC2) G6 instances powered by NVIDIA L4 GPUs are now available in Asia Pacific (Seoul), and Europe (Paris) regions. G6 instances can be used for a wide range of graphics-intensive and machine learning use cases. Customers can use G6 instances for deploying ML models for natural language processing, language translation, video and image analysis, speech recognition, and personalization as well as graphics workloads, such as creating and rendering real-time, cinematic-quality graphics and game streaming. G6 instances feature up to 8 NVIDIA L4 Tensor Core GPUs with 24 GB of memory per GPU and third generation AMD EPYC processors. They also support up to 192 vCPUs, up to 100 Gbps of network bandwidth, and up to 7.52 TB of local NVMe SSD storage. Amazon EC2 G6 instances are available today in the AWS US East (N. Virginia and Ohio) , US West (Oregon), Europe (Paris, Frankfurt, London, Spain, Stockholm and Zurich), Asia Pacific (Seou...

Amazon Connect Tasks can now be configured to expire up to 30 days from creation, with a default of 7 days. For example, less urgent tasks like monthly expense notifications can stay active for up to 30 days before escalating to a supervisor, while urgent tasks like customer escalations can now be sent to a supervisor after 1 minute. Amazon Connect Tasks empowers you to prioritize, assign, and track all contact center agent tasks to completion, improving agent productivity and ensuring customer issues are quickly resolved. This feature is supported in all AWS regions where Amazon Connect is offered. To learn more, see our documentation . To learn more about Amazon Connect, the easy-to-use cloud contact center, visit the Amazon Connect website . Post Updated on December 18, 2024 at 06:00PM

If you haven’t already heard, GitHub just announced Copilot Free ! And great news: you can already start using Copilot Free right now in Visual Studio. It's available now, and we're putting the finishing touches on a more streamlined experience in January. We’re excited to bring these benefits to the entire Visual Studio family. With Copilot Free, you’ll have AI-powered tools right at your fingertips—smarter debugging, AI generated commit messages, AI-breakpoint placement, and of course, direct access to Completions, Chat, and much more. Developers have seen a 25% speed increase with Copilot—and now you can experience that boost firsthand in Visual Studio. Ready to give it a try? [cta-button text=" Sign up for GitHub Copilot Free" url="https://ift.tt/n0SFkDM" color="btn-primary"] A couple of recommendations Try it now: Don’t wait to experience the benefits. Give GitHub Copilot Free a spin in Visual Studio today. Just note that f...

Today, we're announcing the Public Preview of Managed Identities as Federated Identity Credentials for Microsoft Entra. Securely access Entra-protected resources like Microsoft Azure, Microsoft Graph, and third-party APIs using a managed identity instead of a secret or certificate. Key benefits: Improved security: Eliminating the use of secrets and certificates in app authentication reduces the risk of credential leaks. Simplified setup: Using a managed identity as a federated identity credential (FIC) provides continuous access to resources without the need to manage secret and certificate expiration and renewal. How it works In addition to client secrets and certificates, apps today can use Federated Identity Credentials (FICs) to accept access tokens from trusted identity providers. This process, known as the Workload Identity Federation flow, supports tokens from GitHub, Kubernetes, and other third-party OIDC issuers. With this new capability, apps can also accept ...

Some time ago, I showed how to combine two files in git while preserving their line history and how to split a file into two while preserving git line history . Both of these techniques rely on merge commits. But what if your team's policy is to rebase or squash all commits? Can you accomplish these tasks without merge commits? Git's line attribution algorithm follows file history, so let's look at how git tracks file history. To determine the file history connections for a file between a commit and its parent or parents, git looks for the file in each parent commit at the same path. If it's found there, then git considers the file to have been modified in place with respect to that parent. If it's not present in the parent commit at the same path, then git looks to see if the file is similar¹ to a file that is present in the child commit but missing in the parent. If it finds one, then it considers the file to have moved. Otherwise, the file is considered to hav...

Since the release of vector preview, we've been working with many customers that are building AI solution on Azure SQL and SQL Server and one of the most common questions is how to support high-dimensional data, for example more than 2000 dimensions per vector. In fact, at the moment, the vector type supports "only" up to 1998 dimensions for an embedding. One of the impressions that such limitation may give, is that you cannot use the latest and greatest embedding model offered by OpenAI, and also available in Azure, which is the text-3-embedding-large model, as it returns 3072 dimensions. Well, that's not the case. And I would even add that using such high number of dimensions is not really giving you that much benefit compared to the costs that comes with that usage. Let me show you why. Embedding dimensions and MTEB Benchmark " MTEB is a massive benchmark for measuring the performance of text embedding models on diverse embedding tasks. " as stated here...

I ran across some notes I took on a team meeting hosted by a senior executive in what I'm guessing was the late 2010s. As I noted some time ago , I keep an eye out for new jargon terms so I can add them to my Microspeak citations. This is particularly true of meetings with senior executives, because they are often the source of new jargon terms, which then get picked up by lower-level employees, because as my colleague Michael Grier noted, " The intended purpose for this jargon is not to communicate with the people who work for you, but to impress the people you work for ." If a senior executive likes golf and uses a lot of golfing metaphors, you're going to start using golfing metaphors because your boss likes them. My notes for this particular meeting included three new terms, none of which appears to have stuck around, but I record them here for posterity (or in case they make a comeback). Taking on new work puts existing tasks at risk. We need to create airspace...