Augusto Marietti

We are excited to announce the launch of Ursa engine, our Kafka API-compatible data streaming engine that is built on top of Lakehouse. Ursa is not something entirely new; it is built on our many years of experience building and operating #ApachePulsar and #ApacheKafka. To learn more about the #Ursa engine, I recommend our #PulsarSummit keynote presentation and announcement blog post. Here, I’d like to delve deeper into some of its features: 1. Making Kafka API a first-class citizen: In #ApachePulsar, we developed a unified messaging model combining queuing and streaming semantics into one protocol, aimed at enhancing data sharing between business applications. Although powerful and increasingly adopted by enterprises and unicorns, the requirement for application rewrites is impractical in today’s economy. Instead of building data gravity around a protocol, we open up access to the robust stream storage layer by extending support to the Kafka protocol, transitioning from a single protocol to a multi-protocol platform, facilitating greater flexibility and integration. 2. Achieve Stream-Table duality via Lakehouse Storage: Data streams commonly end up in data lakehouses, where integrating and maintaining them can be complex and resource-intensive. Ursa simplifies this by storing data streams as lakehouse tables in object storage. This integration not only streamlines operations but also allows seamless access to operational data for analytical processes. Thanks to its multi-modal storage engine which fosters data sharing across teams and departments. 3. Object storage as the primary storage solution: To better align with cloud economics, particularly for large workloads, Ursa makes ZooKeeper and BookKeeper optional and utilizes a cost-optimized write-ahead log based solely on object storage with leaderless and stateless brokers. This architecture leverages object storage for both data replication and storage, minimizing inter-AZ data transfers and reducing costs for latency-relaxed, high-volume workloads. By integrating these features at the multi-tenant level rather than at isolated cluster levels, Ursa eliminates costly data duplication and movement across clusters, allowing teams to map directly to tenants and be able to share data across teams and departments more efficiently. We believe that the future of data streaming platforms lies in being open and multi-faceted, fundamentally different from traditional data platforms. They are designed not just to store and process data, but to enhance organizational capabilities by fostering data sharing across the organization and integration with other systems. More and more data streaming platforms will look more like Ursa, with multi-protocol, multi-tenant, and multi-modal capabilities. Check out our announcement blog post about Ursa: https://lnkd.in/ghEKTYE6 #ApacheKafka #Lakehouse #DataStreaming

100

1 Comment

New on First Round Capital Review, Will Larson digs into why common management practices have become anti patterns, and which ones engineering execs should lean into to sharpen their team strategy. Always fun to learn from old Stripe co-workers :)

78

1 Comment

This is the initial infrastructure Tradescoop was built on. A few things have changed, some additions and removals; for example, we’ve gone from MongoDB on Digital Ocean to Mongo Atlas, Redis cloud to Memorystore, but the general appraoch remains. Simplicity. Shouts to Karl Obinna Amalu, CSM for cleaning it up and adding color ✨ #gcp #infrastructure #devops

19

1 Comment

When building RAG and LLM apps, the devil is in the documents. One of the hardest and least appreciated pieces of any RAG pipeline is turning complex visual documents into LLM-ready data. I'm talking about financial, health, insurance, gov docs and more. If they have tables, charts, diagrams and so on, it's a recipe for downstream hallucination. So how you fix it? This episode of RAG Masters explains the theory and then looks at PyPDF, Tesseract, LlamaParse, Unstructured and EyeLevel's X-Ray. See who comes out on top. https://lnkd.in/eTJw4hgX

9

"If you're relying on AI not becoming 100X better, you're about to get steamrolled." -- Jason Warner, ex-CTO of Github and CEO of poolside How we code and build software is being rapidly transformed with AI. Sitting down with Jason the future looks clearer. Over the next 5 years: 🌏 All 7 billion people can code 🏗️ Search bars will spit out software 🔋 AI will generate its way out of data scarcity 🌩️ Foundation models to become cloud companies I unpack the 5 year frontier of software development in Future of Code: AI-led Software Development https://lnkd.in/gze2QSbu #AI #Software #Code

10

Anthropic just launched Claude 3.5 Sonnet, which beats both OpenAI GPT-4o and Anthropic’s previous paid model Claude 3 Opus on most benchmarks. Claude 3.5 Sonnet can execute and debug code, and has improved image understanding capabilities, including text transcription. A new feature called “Artifacts” enables users to view generated code, documents and websites in a separate window next to the chat. Claude 3.5 Sonnet is available for free to all users, but paid users get higher usage caps. Did Anthropic just take the lead back from OpenAI? It seems that ChatGPT’s Code Interpreter tool, which used to be one its main advantages, has finally met its match, but ChatGPT still has the advantages of real-time Internet access and custom GPTs which Claude lacks… What do you think? https://lnkd.in/d_hk86cf

10

3 Comments

Big news today... FOSSA free plan just got a major upgrade! We want to make it ridiculously easy for development teams to start securing their software and drive comprehensive Open Source Compliance, Security and Quality -- for FREE. That's why we have now launched a new free forever plan -- built for complete workflows around the lifecycle of managing open source such as: - Open source security - SBOM management - License compliance - Quality / Obsolescence No vendor evals, no sales convos, no zoom meetings. Just sign up, start doing -- and come back if you need to scale. Let's get folks off the dev team's back and ship compliant, secure and high quality software with OSS. Check out the link to the full announcement in the comments.

85

8 Comments

Very useful overview of the CDN market...

3

Introducing the Google Cloud Private Marketplace, now generally available. Private Marketplace allows cloud administrators to create a curated collection of vetted software and solutions that integrate with Google Cloud and align with your organization’s cost, security, and compliance policies. Empower your end users to explore pre-approved tools without compromising on governance and cost control. Read the blog to learn more. #GoogleCloudMarketplace #GCP #GoogleCloud #Marketplace #Software

1

There's a reason why AWS offers so many pricing models. Most companies are doing it wrong. Here are 3 things about Convertible Reserved Instances (CRIs) that will change how you think about AWS cost optimization: When people talk about optimizing AWS costs, they often focus on Reserved Instances (RIs). They think: 1. RIs give the biggest discounts 2. Committing to specific instance types is fine 3. Flexibility isn't worth the trade-off That's old-school thinking, and it's costing you money. The entire approach to AWS cost management is changing. Here's why CRIs are the future (and the future is already here): 1. Flexibility is king RIs are like buying a car with cash. Sure, you get a discount, but you're stuck with it. CRIs are like leasing that car, but with the freedom to swap it for any other vehicle, anytime, while keeping your discount. 2. Adapt at the speed of business Your infrastructure needs change faster than ever. With CRIs, you can: - Adjust capacity on the fly - Switch between instance families - Change OS types and tenancies All while maintaining your cost savings. Try doing that with traditional RIs (you can't). 3. Financial benefits without handcuffs CRIs give you the best of both worlds: significant discounts AND the ability to adapt. In a world where tech moves at lightspeed, why lock yourself into a 3-year commitment? TAKEAWAY Some folks still think that if you want to optimize AWS costs, go all-in on RIs. I think they're talking about maximizing discounts at the expense of flexibility. To me, true optimization looks different... Traditional RIs are dying, cloud needs are evolving rapidly, and flexibility is becoming non-negotiable. To get really fancy with it, you'll need to build a cRI optimization engine that optimizes cRIs in real time - which is exactly what you get with CloudFix.

42

4 Comments

What OpenAI's GPT-4o release means for the future of SaaS -- I'll break down the threat in my new article for Inc. Magazine https://lnkd.in/eUvZ4Ywq #innovation #technology #AI #software #saas

3

Enjoy reading the Datadog blog about how they adopted a new indexing strategy for time-series data, which aligns closely with what GreptimeDB is developing: 1. Written entirely in Rust. 2. Always build inverted indexes for time-series data to avoid full table scans. 3. Shards data and indexes by ranges. 4. Accelerates queries using parallel processing combined with partitioning and indexing. Furthermore, GreptimeDB aims to unify logs, metrics, and events, supporting log-searching capabilities as well. Additionally, all of this data and indexing can be stored on object storage. The Datadog blog is here: https://lnkd.in/gr36NfQr Indexing Techniques in GreptimeDB: https://lnkd.in/gjwp3hHw #Datadog #GreptimeDB #Monitoring #Observability

4

If you are working with Docker... I highly recommend looking at the Slim toolkit. https://lnkd.in/gX_Gb9Kf Why? Slim allows developers to inspect, optimize, and debug their containers using its xray, lint, build, debug, run, images, merge, registry, vulnerability (and other) commands. It simplifies and improves your developer experience building, customizing and using containers. It makes your containers better, smaller and more secure while providing advanced visibility and improved usability working with the original and minified containers. Don't change anything in your container image; minify it by up to 30x to make it secure, too! Optimizing images isn't the only thing it can do, though. It can also help you understand and author better container images. Keep doing what you are doing. No need to change anything. Use the base image you want. Use the package manager you want. Don't worry about hand-optimizing your Dockerfile. You shouldn't have to throw away your tools and your workflow to have small container images. Node.js application images: from ubuntu:14.04 - 432MB => 14MB (minified by 30.85X) from debian:jessie - 406MB => 25.1MB (minified by 16.21X) from node:alpine - 66.7MB => 34.7MB (minified by 1.92X) from node:distroless - 72.7MB => 39.7MB (minified by 1.83X) Go application images: from golang:latest - 700MB => 1.56MB (minified by 448.76X) from ubuntu:14.04 - 531MB => 1.87MB (minified by 284.10X) from golang:alpine - 258MB => 1.56MB (minified by 165.61X) from centos:7 - 615MB => 1.87MB (minified by 329.14X) Rust application images: from rust:1.31 - 2GB => 14MB (minified by 147.16X) Java application images: from ubuntu:14.04 - 743.6 MB => 100.3 MB

2

Big congrats to Eugene Cheah and the Recursal AI team for the release of Featherless AI, which enables serverless inference of every Llama-3 8B and 70B model on Hugging Face That’s over 475 models w/ more being added daily, thereby enabling anyone to quickly experiment, try, and choose the latest and best models, from Hugging Face, starting from $10/month. To use even the smallest fine-tuned models requires dedicated hardware, which translates to real hosting costs, whether you’re experimenting with a model or ramping up production use. The goal of featherless is to make every model on Hugging Face available serverless and with these Llama & RWKV based models, Recursive AI has taken a big step of the way there.

18

3 Comments

Datadog's Annual Conference 𝐃𝐀𝐒𝐇 𝟐𝟎𝟐𝟒: 𝐍𝐞𝐰𝐞𝐬𝐭 𝐚𝐧𝐧𝐨𝐮𝐧𝐜𝐞𝐦𝐞𝐧𝐭𝐬 At this year’s DASH, we announced new products and features that enable your team to 𝗼𝗯𝘀𝗲𝗿𝘃𝗲 your environment, 𝘀𝗲𝗰𝘂𝗿𝗲 your infrastructure and workloads, and 𝗮𝗰𝘁 to remediate problems before they affect customers. One of the announcements I'm personally most excited about is that 𝗟𝗟𝗠 𝗢𝗯𝘀𝗲𝗿𝘃𝗮𝗯𝗶𝗹𝗶𝘁𝘆, which enables you to get deep visibility into your generative AI applications, is now generally available! Explore the full blog post to discover all the updates announced at this year's DASH. #Dash2024 #Datadog #Observability #Security #Announcement #LLM #GenAI

29

I am a big fan of RED Method. Love how Coroot makes it easy to apply it to Distributed Tracing Data https://lnkd.in/ezG84KtX #coroot #otel #observability

6

Great SRE Interview Questions 

4

If the CDN market will grow 3%, it may mean the market has STOPPED GROWING (~3% is rate of inflation). That seems like a big story, and maybe an indicator that the internet is at a pivot from stored file delivery (CDNs, video streaming) to more real-time traffic and compute.

31

8 Comments