.png){kind=icon}
.png){kind=icon}
Return
Adaptive Gradient Coding
.png){kind=icon}
OA
AI
DOI：10.1109/TNET.2021.3122873.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper focuses on mitigating the impact of stragglers in distributed learning system. Unlike the existing results designated for a fixed number of stragglers, we develop a new scheme called Adaptive Gradient Coding (AGC) with flexible communication cost for varying number of stragglers. Our scheme gives an optimal tradeoff between computation load, straggler tolerance and communication cost by allowing workers to send multiple signals sequentially to the master. In particular, it can minimize the communication cost according to the unknown real-time number of stragglers in practical environments. In addition, we present a Group AGC (G-AGC) by combining the group idea with AGC to resist more stragglers in some situations. The numerical and simulation results demonstrate that our adaptive schemes can achieve the smallest average running time.
Keywords:
Encoding
Costs
Codes
Real-time systems
Task analysis
Optimization
Adaptive systems
Gradient coding
straggler
adaptive
distributed computing
Journal
I
IF:
3.6
Papers:
4.4K
Citations:
9.5K