Finding training inefficiencies with CentML DeepView
41 min

Finding training inefficiencies with CentML DeepView

MLOps World - MLOps World & Generative AI World 2024
Yubo Gao
visual-profiler training-throughput batch-size mixed-precision tensor-cores flash-attention deep-learning machine-learning performance-optimization profiling gpu-utilization pytorch tensorflow model-deployment cost-optimization energy-efficiency
Lessons learned from scaling large language models in production
41 min

Lessons learned from scaling large language models in production

MLOps World - MLOps World & Generative AI World 2024
Matt Squire
ray-serve large-language-models llm rag mlops gpu performance-optimization inference scaling python fastapi kubernetes vm vector-database
Finding the Hidden Drivers of AI Business Value
10 min

Finding the Hidden Drivers of AI Business Value

MLOps World - MLOps World & Generative AI World 2024
Jakob Frick
generative-ai ai-impact business-value mlops ml-in-production llm frameworks value-creation ai-products performance-optimization
Continuous Profiling Your Production Kubernetes Deployments Running at Scale
29 min

Continuous Profiling Your Production Kubernetes Deployments Running at Scale

KubeCon + CloudNativeCon - KubeCon + CloudNativeCon Europe 2023
Vijay Samuel Nick Pordash
kubernetes continuous-profiling observability profiling go performance-optimization cpu-profiling memory-profiling open-telemetry pyroscope docker ci-cd
Best Practices for Accelerated Image Distribution Using Dragonfly
29 min

Best Practices for Accelerated Image Distribution Using Dragonfly

KubeCon + CloudNativeCon - KubeCon + CloudNativeCon Europe 2023
Wenbo Qi Yiyang Huang
dragonfly p2p image-distribution container-images kubernetes cloud-native performance-optimization registry storage networking ci-cd
Understand Systems with OpenTelemetry: A Hybrid Telemetry Data Backend
16 min

Understand Systems with OpenTelemetry: A Hybrid Telemetry Data Backend

KubeCon + CloudNativeCon - KubeCon + CloudNativeCon Europe 2023
Ran Xu Xiaochun Yang
opentelemetry observability telemetry backend-storage cloud-native distributed-systems data-management performance-optimization huawei-cloud
You Don't Have to Run it Locally! How to run your emulators in the cloud.
26 min

You Don't Have to Run it Locally! How to run your emulators in the cloud.

Droidcon - Droidcon London 2024
Petras Vičiūnas
android android-emulator cloud-computing kubernetes docker ci-cd remote-development debugging performance-optimization virtualization webrtc screen-sharing
Jetpack Compose: Drawing without pain and recomposition
40 min

Jetpack Compose: Drawing without pain and recomposition

Droidcon - Droidcon London 2024
Vitalii Markus
jetpack-compose android-development ui-toolkit performance-optimization recomposition drawing memory-management declarative-ui kotlin modifier android
Improving App Performance With Custom Thread Pools
12 min

Improving App Performance With Custom Thread Pools

Droidcon - Droidcon London 2024
Tin Novakovic
android-development app-performance thread-pools coroutines ok-http room-persistence-library concurrency multithreading performance-optimization java kotlin android
From realtime CameraX filters to General Purpose GPU computing in simple steps
31 min

From realtime CameraX filters to General Purpose GPU computing in simple steps

Droidcon - Droidcon London 2024
Konstantin Zolotov
gpu-computing camerax android-development opengl glsl compute-shaders fragment-shaders vertex-shaders image-processing realtime-rendering performance-optimization
Oh no! My regex is causing a Denial of Service! What can I do about it?
31 min

Oh no! My regex is causing a Denial of Service! What can I do about it?

PyCon - PyCon US 2023
Guillaume Dequenne Quazi Nafiul Islam
regex regular-expressions redos denial-of-service performance-optimization static-analysis python backtracking
How Memory Profilers Work
31 min

How Memory Profilers Work

PyCon - PyCon US 2023
Pablo Galindo Salgado
python memory-profiler memory-leaks performance-optimization data-transformation code-optimization memory-management allocation-patterns