From 88ad1d97fa51db59ce4ce256178d03b137a64df0 Mon Sep 17 00:00:00 2001 From: Yoav Dobrin Date: Tue, 30 Jan 2024 16:05:29 +0200 Subject: [PATCH 1/2] decrease font size, removing redundant file --- _sass/color_schemes/custom.scss | 4 +-- content pack proposal.md | 61 --------------------------------- 2 files changed, 2 insertions(+), 63 deletions(-) delete mode 100644 content pack proposal.md diff --git a/_sass/color_schemes/custom.scss b/_sass/color_schemes/custom.scss index a8177b3..2a53dcc 100644 --- a/_sass/color_schemes/custom.scss +++ b/_sass/color_schemes/custom.scss @@ -58,7 +58,7 @@ $link-color: $blue-000; li { width: 50%; - height: auto; + height: 60px; justify-content: center; display: flex; list-style: disc; @@ -81,7 +81,7 @@ $link-color: $blue-000; flex-direction: column; .text { - font-size: 1.5rem; + font-size: 1.2rem; } } } diff --git a/content pack proposal.md b/content pack proposal.md deleted file mode 100644 index cd9bbb4..0000000 --- a/content pack proposal.md +++ /dev/null @@ -1,61 +0,0 @@ -# Content Pack Proposal - -ISV and DN customers are transitioning from POC/MVP to production deployments. Transitioning a POC into a production environment requires careful attention to a multitude of crucial factors: - -- Performance -- Networking & Security -- Cost Management/Optimization -- Reliability/High Availability -- Scaling & Quota Management -- Observability -- Multi-tenant Architecture - -While there is an abundance of resources available online, our objective is to consolidate this information into a comprehensive, easy-to-follow guide. Where gaps in the information exist, we will try to fill them by creating additional articles or providing sample code. We propose the following new content. - -## Performance Evaluation Tools for AI Systems - -Proposal: We suggest creating a blog post or article that provides a comprehensive guide on performance testing for AI applications. This piece will offer insights into conducting load testing, assessing system efficiency - including latency and stability - and specifically evaluating the performance of the AI model, which is the core of the application. - -### Existing content - -[https://www.microsoft.com/en-us/research/blog/assessing-ai-system-performance-thinking-beyond-models-to-deployment-contexts/](https://www.microsoft.com/research/blog/assessing-ai-system-performance-thinking-beyond-models-to-deployment-contexts/) - -## Strategies for Managing High Volume Token Demand - -Proposal 1: We propose an article designed to aid in cost reduction, latency minimization, and performance enhancement when using LLMs. It will also provide guidance on quota management, particularly in scenarios where a high volume of tokens are utilized. This practical guide will serve as a valuable tool for developers seeking to optimize their applications. - -Proposal 2: We propose to develop code samples and architecture designs for implementing high volume token solutions - -### Existing content - -1. [Token Optimization: Efficient AI Conversations with OpenAI](https://techcommunity.microsoft.com/t5/healthcare-and-life-sciences/unlocking-the-power-of-tokens-optimizing-token-usage-in-gpt-for/ba-p/3826665) -1. [Azure OpenAI Service Load Balancing with Azure API Management](https://learn.microsoft.com/samples/azure-samples/azure-openai-apim-load-balancing/azure-openai-service-load-balancing-with-azure-api-management/) - -## Insights into Embedding Upgrades and Version Control in AI - -Proposal: We propose creating a blog post or article, complemented by an architectural diagram, that focuses on adapting to changes in embedding models. This resource will specifically address scenarios that necessitate the recalculation of embeddings. The aim is to provide a clear, visual guide to help understand and effectively manage changes in their embedding models. - -1. Azure OpenAI embedding model updates. - 1. Scenarios where input-output dimensions remaining the same but improved models. - 1. Scenarios where input-output dimensions also changes with model improvements. -1. Customer wants to replace the AOAI embedding models with open-source ones. - -## Approach on how to handle zone/region and Failover - -Proposal: We suggest developing a blog post or article, as well as code samples, that outlines strategies for handling zone/region management and failover scenarios. This resource will focus on the creation of resilient applications, providing practical guidance and hands-on examples to build robust and fault-tolerant systems. - -## Schematic Representation of AI System Monitoring Frameworks - -Proposal: Code Snippets for Effective AI System Monitoring - -### Existing content - -[Enterprise Azure OpenAI Logging](https://github.com/Azure-Samples/openai-python-enterprise-logging) - -## Guide to Monitoring AI Systems: Limits, Quota, and Health - -Proposal: article that highlights observability metrics to look for, while working with AI systems - direct impact on performance , cost etc - -### Existing content - -[Implement logging and monitoring for Azure OpenAI models](https://learn.microsoft.com//azure/architecture/ai-ml/openai/architecture/log-monitor-azure-openai) From b3e7329ff1ddbcc4fc89f95f685d27edeb6d4bad Mon Sep 17 00:00:00 2001 From: Yoav Dobrin Date: Tue, 30 Jan 2024 17:40:09 +0200 Subject: [PATCH 2/2] links open in new tab --- _includes/head_custom.html | 1 + _sass/color_schemes/custom.scss | 1 - assets/js/open-links-in-new-tab.js | 10 ++++++++++ 3 files changed, 11 insertions(+), 1 deletion(-) create mode 100644 _includes/head_custom.html create mode 100644 assets/js/open-links-in-new-tab.js diff --git a/_includes/head_custom.html b/_includes/head_custom.html new file mode 100644 index 0000000..525891f --- /dev/null +++ b/_includes/head_custom.html @@ -0,0 +1 @@ + \ No newline at end of file diff --git a/_sass/color_schemes/custom.scss b/_sass/color_schemes/custom.scss index 2a53dcc..4bee7b1 100644 --- a/_sass/color_schemes/custom.scss +++ b/_sass/color_schemes/custom.scss @@ -58,7 +58,6 @@ $link-color: $blue-000; li { width: 50%; - height: 60px; justify-content: center; display: flex; list-style: disc; diff --git a/assets/js/open-links-in-new-tab.js b/assets/js/open-links-in-new-tab.js new file mode 100644 index 0000000..1c6a60a --- /dev/null +++ b/assets/js/open-links-in-new-tab.js @@ -0,0 +1,10 @@ +document.addEventListener("DOMContentLoaded", function() { + var links = document.querySelectorAll("a[href]"); + links.forEach(function(link) { + if (link.href.startsWith("http") && !link.href.startsWith(window.location.origin)) { + link.target = "_blank"; + link.rel = "noopener noreferrer"; // For security and performance reasons + } + }); + }); + \ No newline at end of file