Hongchao Deng

Build a real-time lakehouse with Apache Doris + AWS Glue + S3 Tables ➡️ Join our meetup in SF to see the full demo: https://lnkd.in/gRmeZWUd 🔧 Stack: - S3 stores Iceberg table data - AWS Glue as a central metadata catalog for schema and table management - Apache Doris (via VeloDB) brings unified batch + real-time SQL analytics on top 💡 Pros of this setup: - Real-time analytics: Doris delivers sub-second, high-concurrency queries directly on S3 data. - Open architecture: Keep data in open formats (Parquet, Iceberg) for multi-engines access. - Simple AWS integration: Glue makes tables easily managable across AWS services (Athena, EMR, Redshift Spectrum...) - Cost-efficient: Cheap S3 storage, elastic Doris compute, and no complex ETL pipelines. Our quick demo below shows connecting Glue and S3 to Doris through VeloDB Studio, then create and query a new table, all in real time. 🎟️ RSVP link at the top, join our event in SF to see our demo and demos from Uber, Pinterest, and Datastrato. #AWS, #ApacheDoris, #Gravitino

13

We’re excited to see this new tutorial from our partners at dstack, showing how to seamlessly connect to Crusoe’s high-performance GPU clusters for your model training and inferencing jobs. Whether you prefer the flexibility of using Crusoe Managed Kubernetes or direct access via Crusoe GPU Virtual Machines, this integration helps you automate cluster provisioning and management for model training at scale. Check out the full guide to see how you can leverage our powerful Crusoe Cloud for your model training. 👇

69

1 Comment

In this post, we demonstrate how to integrate Amazon SageMaker HyperPod with Anyscale platform to address critical infrastructure challenges in building and deploying large-scale AI models. The combined solution provides robust infrastructure for...

1

"Pull all siloed AI projects into a one-page chart. Focus on the ones aligned with strategic goals, pause the rest, assign owners, remove roadblocks, and make clear what business value each delivers." – Julian Tang, AI Champion, BlackRock In this Pure Storage–sponsored episode of the #AIinBusinessPodcast, Julian Tang joins Emerj Artificial Intelligence Research Editorial Director Matthew DeMello to break down how enterprises can operationalize AI, align projects with strategy, and accelerate measurable outcomes. Retail and CPG leaders can uncover actionable frameworks to integrate AI into workflows, prioritize initiatives, and unlock ROI with greater clarity and speed. Listen to the full episode to discover actionable strategies for aligning AI projects with business goals and accelerating real results: https://lnkd.in/dxxHiDGU #EnterpriseAI #AIGovernance #AIWorkflow

3

1 Comment

Ever wondered how distributed applications can get smarter about scheduling tasks? Google Cloud has partnered with Anyscale to integrate Ray—an open-source compute engine—and Kubernetes via its GKE. Why is this a breakthrough? Imagine: 💡 Assigning labels like `gpu-family=L4` to Ray nodes. 💬 Configuring fallback logic for GPU availability and spot markets. 💪 Integrating Jax frameworks. - The Plus? Full scheduling control ensures no gaps. For details visit: https://lnkd.in/d_jd5cQn Find detailed instructions, capabilities and demo guides covering TPU, GPE, or Kubernetes-based lightweight assignment capable performance engines.

1

Discover how the #OpenTelemetry Operator automates telemetry data collection in Kubernetes, streamlining your #observability practices. Engineer Israel Blancas details how the operator works with #Coralogix out of the box and provides a five-step tutorial on how to get started: https://lnkd.in/dhhYQ2jU.

20

Deploying observability should be seamless—not a coding slog 🔧 Coralogix’s guide on using the OpenTelemetry Operator shows how to set up full tracing with zero code changes—just a Helm chart! In minutes you can auto-collect trace data from supported languages, stream it directly into Coralogix, and start analyzing in real-time. No sidecars, no manual instrumentation—just powerful insights fast. This makes it a breeze to scale visibility across Kubernetes environments and empowers teams to go from deploying to diagnosing without delay. With auto-instrumentation handled transparently, developers can focus on innovation, not overhead. Ready in minutes—not days 🚀 #observability #OpenTelemetry #Kubernetes #DevOps #Coralogix #tracing

2

🚀In the #DSS2025, Peter (Zhenqiu) Huang from Uber and Shiyan Xu from Onehouse dive deep into how #Uber is modernizing its Cloudlake with #ApacheFlink Streaming Ingestion and #ApacheHudi—achieving fresher data, simpler pipelines, and better cost efficiency. 🌟 Highlights include: ✅ Why real-time ML requires <15-minute freshness ✅ Flink vs. batch ingestion: cost, efficiency, and architecture ✅ Key tradeoffs in commit size, sorting, and file handling ✅ Autoscaling + DR for large-scale streaming pipelines ✅ How Hudi’s MOR tables unlock high-throughput streaming ▶️ Watch the recording now!
https://hubs.ly/Q03XZq6Y0 #DataStreaming #RealTimeData #DSS #datastreamingsummit

8

If you’ve seen my explainer video on MCP and would now like to see actual practical things you can do with it, _watch this video_. Sami walks you through using the Confluent MCP server to do things in Confluent Cloud under voice control. Now imagine taking this demo and extending it to run from your own agentic microservice. This stuff is all very, very doable now…

62

1 Comment

🎉 NVIDIA Dynamo V0.2 now delivers GPU autoscaling, Kubernetes support automation, and networking optimizations. ➡️ https://nvda.ws/43FwfLS The latest release includes: ✅ A planner for prefill and decode GPU autoscaling ✅ Kubernetes automation for large-scale Dynamo deployments ✅ Support for Amazon EFA for internode data transfers on Amazon Web Services (AWS) Dynamo is a high-throughput, low-latency #opensource inference serving framework for deploying generative AI and reasoning models in large-scale distributed environments. #COMPUTEX

155

1 Comment

▶️ In our "Getting to Grips with Confluent" video series, we have explored stream processing, Apache Kafka, and core Confluent mechanics. In our fourth video, Oscar Moores compares Confluent Platform and Confluent Cloud. While both services deliver similar core functionality, they differ in ease of use, management, and operational requirements. Confluent Cloud requires no physical infrastructure and comes with a managed schema registry and managed connectors, making it simpler for many organisations to deploy. 🔗 Find out more when you watch the full episode: https://lnkd.in/eH34nMRs Ozan Guzeldereli #Confluent #CloudSolutions #Kafka

3

Diffblue is now a GitHub Copilot launch partner. GitHub just announced custom agents for Copilot and they selected Diffblue alongside HashiCorp, Dynatrace, JFrog, MongoDB, and PagerDuty to bring specialized AI capabilities directly into the developer workflow. What this means for enterprise Java testing: - Native distribution across GitHub's 100M+ developer ecosystem - Integrated into VS Code, CLI, and https://lnkd.in/g478SgzT - The deterministic alternative to probabilistic unit test generation methods As AI coding assistants become standard, the need for reliable, verifiable test automation only grows. We're excited to be part of this ecosystem. Read more: https://lnkd.in/gt4_qhTQ

47

.@AWSCloud fires up Project Rainier, Trainium2 cluster for @AnthropicAI http://bit.ly/4o8ou96 Amazon Web Services said Project Rainier, an AI compute cluster powered by 500,000 Trainium2 chips, is now in use for Anthropic.

“1 million rows per second.” In his opening keynote, our CEO Max Liu walked through what that really means in production: - TiDB processing 1M rows/sec - 10x higher throughput for cross-region replication - A data layer built to make multi-tenant architectures actually practical at scale If you’re thinking about SaaS, multi-region, or AI-heavy workloads, this is the kind of infrastructure ceiling you want. 🎥 Watch the full keynote from Max Liu at #TiDBSCaiLE to see what’s behind the numbers: https://ow.ly/BgKl50YetBq #TiDB #DistributedSQL #MultiTenancy #CloudNative #Scalability

25