Aparna Subramanian

Thanks again for having me at #SUGJaipur. The recording of my talk: "XM Cloud Components - what's new?!" is up on their #youtube and available here: https://lnkd.in/dkVbARfP

21

YugabyteDB ❤️ Apache Airflow: check out this talk at the Airflow Summit by Amit Chauhan.

35

Tools like #Kubernetes can incur significant costs if not properly configured or managed. Harness CEO Jyoti Bansal has a three-prong approach for effectively managing costs: 1️⃣: Visibility 2️⃣: Optimization 3️⃣: Cost governance See more of Jyoti's thoughts on managing rising costs in this Runtime news roundtable below.

5

The story of MongoDB in India is incredible and I am always inspired by the potential it holds. India's developer community, nearly four million strong, is a powerhouse of talent and creativity and it is no surprise that about 20% of Fortune 500 companies have development facilities here. We are already empowering both established corporations and emerging startups in the country, from supporting Tata Group Digital's super app Tata Neu to helping Zomato manage high-volume data even during peak periods. Hear from Sachin Chawla, our Vice President for India, who spoke with CNBC about our commitment to helping Indian developers and organisations build modern applications and take advantage of the AI revolution.🇮🇳👏🏻 Sachin Shinde, Vivek Ganesh, Aditya Nanda, Bhavesh Punjabi #MongoDB #India #innovation

84

1 Comment

A really nice read from Sai Maddali on the benefits of Tiered Storage with Amazon MSK. https://lnkd.in/gGNJ5nyt #ApacheKafka #Kafka #AWS #MSK #tieredstorage #streaming #storage

55

2 Comments

AI can follow your instructions but still cannot debug an issue as well as you can. https://lnkd.in/g2NddYZH #AI #Observability #SoftwareEngineering by Siddarth Jain

6

Best TechStack used by Netflix Netflix, with over 247.15 million subscribers, delivers a top-notch streaming experience through its meticulously crafted architecture. Designed to be scalable, reliable, and efficient, Netflix's architecture operates on a cloud computing platform, enabling seamless resource management. Leveraging open-source technologies keeps costs in check and ensures future readiness. Central to Netflix's architecture are microservices—small, autonomous units that scale and deploy independently. This flexibility ensures robustness and adaptability. Netflix employs diverse caching methods to lighten server loads. A prime example is its content delivery network (CDN), caching frequently accessed content closer to users for enhanced performance and cost efficiency. Underneath the hood, Netflix's architecture is powered by a suite of technologies: Frontend: React, GraphQL Backend: Spring Boot, Zuul, Eureka Data Storage: EVcache, Cassandra, CockroachDB, S3 (for static asset storage) Messaging/Streaming: Apache Kafka, Flink, RabbitMQ Data Analytics & Personalization: Apache Spark, Python, TensorFlow (for recommendation systems and machine learning) Cloud Computing: Amazon Web Services (AWS), Spinnaker (for continuous delivery), Titus (container management) Observability: Atlas (for telemetry data collection), Chaos Monkey (for resilience testing) This sophisticated system evolves continuously to incorporate new features and accommodate expansion into new markets while upholding core tenets of scalability, reliability, and efficiency. What are your thoughts on Netflix's architecture? 👋 PS - Join and get the powerful system design template (it's free): https://lnkd.in/gmichq_H

725

32 Comments

The topic of "Why Kubernetes" has been a discussion point constantly. To get the ground reality, Kalpesh Bhalekar & I have had multiple interviews across Startups and Mid-Market companies who gave multiple reasons for moving to K8s: 1. High up-time requirement in markets like E-commerce, Travel & Healthcare. 2. Single powerful API to cover AWS, Azure & GCP. 3. Cost efficiency with effective resource utilization. Here are a few reasons why Kubernetes is becoming a household name across the market. If you plan to evaluate Kubernetes, just ping me or sign up for our waitlist. (Link in comments) #kubernetes #devops #developerexperience #cloudtechnology

14

1 Comment

Looking forward to this lightning talk on #serverless with fellow serverless heroes from India Jones Zachariah Noel N Dhaval Nagar Srushith R Four of us are going to look at serverless with our own unique lens, with our own individual experiences and bring them all together in this talk. I’ll be focusing on how #GenAI and #Serverless can come together in building and operating applications, also the point of focus lately at AntStack #BelieveInServerless https://lnkd.in/gvvNnta3

34

ScaleOps - Cloud-Native Optimization just turned up the heat in cloud optimization, grabbing a cool $58M in Series B like it was a no-brainer. Hats off to Yodar Shafrir and Guy Baron—the brains behind this real-time cloud orchestration powerhouse. Lightspeed led this round, with NFX, Glilot Capital Partners, and Picture Capital jumping on board. If you’re not paying attention to what these guys are building, you’re about to get left in the cloud dust. ScaleOps isn’t just fine-tuning the cloud; they’re ripping out the old wiring and rewiring the whole system with AI-powered efficiency. Let’s talk tech—predictive horizontal pod auto-scaling. It’s like handing your cloud the ability to read the future, scaling on command with zero wasted spend. The platform’s smart enough to handle intelligent pod placement like it’s playing 4D chess, lining up resources with precision so you’re not paying for an ounce more than you need. And here’s where it gets wild: ScaleOps cracked the code on automated pod rightsizing, dumping the whole tedious manual game of CPU and memory allocation. It’s not a tweak; it’s a full-blown transformation, letting your DevOps sleep while ScaleOps does the heavy lifting. Meanwhile, their node optimization chops are saving companies thousands in cloud costs with every new deployment. It’s no shock that ScaleOps has #Fortune100 like Wiz, SentinelOne, and Orca Security already in the mix. For anyone serious about squeezing every last drop of efficiency from their cloud budget, this is it. ScaleOps isn’t just in the optimization business—they’re taking the whole category up a notch. So what’s next for these cloud kings? With $58M in the tank, they’re scaling their dev team, building out new features, and expanding operations to make cloud management feel like child’s play. ScaleOps isn’t here to play nice with cloud spend; they’re here to redefine the game. Watch this space, because Yodar and Guy aren’t stopping until they’ve changed the whole damn landscape. #Cloud #CloudComputing #CPU #CloudCost #CloudOptimization #Optimization #AutoScaling #AI #CloudAI #Technology #Innovation #TechEcosystem #StartupEcosystem #VentureCapital #StartupCapital #startups

2

Citius ! Loyal Guru got 2x faster with AlloyDB for PostgreSQL. "Prior to adopting AlloyDB, large traffic spikes from major retail events on Loyal Guru often created bottlenecks. We struggled to deal with data growth, which strained the limits of our storage capacity. AlloyDB's autopilot feature for managing storage has been instrumental in mitigating these issues, automatically adjusting storage capacity to meet our demands without manual intervention. Overall, we reduced query latency by 40-50% and optimized storage from 2TB to 1.3TB — a 35% improvement." #alloydb #gcp #postgresql

65

Huge thanks to the Tavily team 😁 for sharing my tutorial on combining Elasticsearch, LangGraph, and Llama3 for powerful RAG workflows! 🙏 It's exciting to see this setup getting traction, especially with Tavily’s integration for enhanced web search. Together, we’re pushing the boundaries of what’s possible in scalable, efficient, and accurate retrieval-augmented generation (RAG) workflows.🚀 #Elasticsearch #LangGraph #Llama3 #TavilySearch #RAG #AI #MachineLearning #DataScience

29

🚀 November Meetup Alert – Don’t Miss Out! When: November 23rd Where: OpsTree premises Time: 10:00 AM - 2:00 PM Get ready for an action-packed session where you'll learn: ✨ How to build a data streaming platform: Dive into Stream Processing and Stream Governance with Kafka and Flink. 💡 Create a Real-Time Data Warehousing Solution: Explore cutting-edge solutions using AWS and Confluent Cloud. Join us and transform your knowledge with these exciting, hands-on insights. See you there!"

67

A few days back, one of our customers escalated an urgent issue Customer : Urgent! Please help, data flow has stalled Syncari Dev : Let me quickly look at debug logs …and 15 mins in, we had a resolution! We all know, debug logs are god send for developers. Every software runs into issues and the successful ones have the right tools to help quickly identify the issue. This is no different for any data automation tool. As data flows through systems, it’s crucial to identify the root of any discrepancies quickly to avoid data corruption To empower our users on troubleshooting, we released pipeline debug logs.   Read more here https://lnkd.in/gJWHQygP. #mdm #syncari #debug #developers #dataautomation #dataquality

29

2 Comments

Here at nOps, we’re big fans of Karpenter — and here’s why Karpenter + nOps is even better. nOps Compute Copilot enhances Karpenter to make it simple for users to maintain stability and optimize resources efficiently. nOps helps engineering teams more easily leverage the power of Karpenter to save with Spot, optimize purchase commitments, and efficiently provision resources for higher performance at lower costs. nOps adds the following capabilities to Karpenter — read all about it here: https://lnkd.in/dQza83Jm #karpenter #nops #awsoptimization

16

Very interesting post on Intuit's use of LLMs to manage K8s, via Aditya Kulkarni at InfoQ: "Intuit Engineering's Approach to Simplifying Kubernetes Management with AI" https://lnkd.in/eRN6_sAK

2

Amazon Managed Service for Apache Flink now supports Amazon DynamoDB Streams as a source! The new connector, contributed by #AWS for the #ApacheFlink #opensource project, adds Amazon #DynamoDB Streams as a new source for Apache Flink. You can now process DynamoDB streams events with Apache Flink, a popular framework and engine for processing and analyzing #streamingdata. https://lnkd.in/gDFsiXA2

17

Use Flux and SpinKube together! Check out this latest blog post based on the conversation Sohan Maheshwar had with Flux expert Kingdon Barrett https://lnkd.in/gbGwsddq #spin #wasm #flux #cicd

11