NUMA Courses and Certifications

Devoxx

Java 9 - The Quest for Very Large Heaps

Explore Java 9's garbage collection improvements for large memory applications, including G1 architecture, concurrent marking, and enhanced performance for data-intensive software.

• YouTube
• 52 minutes
• On-Demand
• Conference Talk

Under the Hood of a Shard-per-Core Database Architecture

Explore ScyllaDB's shard-per-core architecture and Seastar framework, learning how modern database design optimizes performance, scalability, and resource utilization in server technology.

• YouTube
• 47 minutes
• On-Demand
• Free Video

Comparing Performance of NVMe Hard Drives in KVM, Baremetal, and Docker - Using Fio and SPDK

Explore NVMe drive performance across KVM, Baremetal, and Docker environments using Fio and SPDK. Learn testing methods and analysis for optimizing virtual testbed applications.

• YouTube
• 43 minutes
• On-Demand
• Free Video

Comparing Performance of NVMe Hard Drives in KVM, Baremetal, and Docker - Virtual Testbed Applications

Explore NVMe drive performance across KVM, Baremetal, and Docker environments using Fio and SPDK. Learn testing methods, tools, and protocols for optimizing virtual testbed applications.

• YouTube
• 43 minutes
• On-Demand
• Free Video

Linux Memory Management and Containers - Tutorial

Explore Linux memory management principles and their impact on containerized processes, covering demand paging, swapping, OOM killing, and NUMA configurations. Practical exercises included.

• YouTube
• 1 hour 36 minutes
• On-Demand
• Free Video

Kubernetes WG Batch: Recent Improvements and Future Roadmap

Explore recent improvements and future plans for batch workload support in Kubernetes, including Job API enhancements, Kueue developments, and hardware resource management advancements.

• YouTube
• 40 minutes
• On-Demand
• Free Video

Linux Plumbers Conference

XDP 1.5 Years in Production - Evolution and Lessons Learned

Evolution of XDP and BPF in production: performance improvements, operational challenges, and future enhancements. Insights from Facebook's network processing implementation using XDP for over 1.5 years.

• YouTube
• 33 minutes
• On-Demand
• Conference Talk

Deep Dive - SIG VMware

Explore Kubernetes scheduling on vSphere, optimizing performance and availability through configuration tuning. Learn to integrate topology labels for improved resource management and high availability.

• YouTube
• 45 minutes
• On-Demand
• Conference Talk

Inside Livermore Lab

RAJA Performance Suite and Umpire Tutorial - 2024

Explore RAJA and Umpire libraries for HPC application portability and memory management across diverse architectures, including NUMA and GPUs.

• YouTube
• 1 hour 34 minutes
• On-Demand
• Free Video

Linux Plumbers Conference

TAO: THP Allocator Optimizations

Explore THP Allocator Optimizations with ZHAO Yu, delving into advanced Linux memory management techniques and performance enhancements.

• YouTube
• 32 minutes
• On-Demand
• Free Video

Leverage Topology Modeling and Topology-Aware Scheduling to Accelerate LLM Training

Explore topology modeling and scheduling techniques to optimize LLM training performance. Learn to address network bottlenecks and enhance resource allocation for AI workloads in Kubernetes environments.

• YouTube
• 45 minutes
• On-Demand
• Free Video

USENIX

UniMem - Redesigning Disaggregated Memory within a Unified Local-Remote Memory Hierarchy

Explores UniMem, a novel disaggregated memory system that optimizes remote memory access, local cache utilization, and page migration for improved performance in large-scale computing environments.

• YouTube
• 20 minutes
• On-Demand
• Free Video

SNIAVideo

Riding the Long Tail of Optane's Comet - Emerging Memories, CXL, UCIe, and More

Explore the post-Optane landscape of emerging memory technologies, examining alternatives, legacy impacts, and new protocols like CXL and UCIe while understanding their role in future computing architectures.

• YouTube
• 46 minutes
• On-Demand
• Free Video

KVM Forum

Exploring VM Placement Strategies for Chiplet Architectures

Discover advanced VM scheduling techniques for chiplet-based processors, optimizing placement strategies to minimize inter-chiplet communication penalties and maximize performance.

• YouTube
• 28 minutes
• On-Demand
• Free Video

USENIX

Quake - Adaptive Indexing for Vector Search

Discover Quake, an adaptive indexing system for vector search that maintains low latency and high recall under dynamic workloads with multi-level partitioning and NUMA-aware parallelism.

• YouTube
• 15 minutes
• On-Demand
• Free Video