Recently, Hama core committers Suraj Menon and Thomas Jungblut are working on Fault-tolerant BSP system. And I am trying to read the source code. Their design describes the new BSP computing system and API enabling checkpoint-based recovery. Furthermore, describes the confined recovery, which can be used to improve the cost and latency of recovery.
I didn't fully understand and test yet but quite nice!
Subscribe to:
Post Comments (Atom)
-
Opening the black box of Deep Neural Networks via Information - https://arxiv.org/pdf/1703.00810.pdf 지금까지 딥 러닝은 어떻게 동작하는지 이해할 수 없다고 믿어져왔다...
-
음성 인공지능 분야에서 스타트업이 생각해볼 수 있는 전략은 아마 다음과 같이 3가지 정도가 있을 것이다: 독자적 Vertical 음성 인공지능 Application 구축 기 음성 플랫폼을 활용한 B2B2C 형태의 비지니스 구축 기 음성 플랫폼...
-
As mentioned ago, I've been forming up the Hamburg project with Hyunsik Choi. Let's see more detail in the diagram of computing met...
No comments:
Post a Comment