Differences

This shows you the differences between two versions of the page.

Link to this comparison view

seminars:adaptive_load_balancer_for_distributed_fpga_acceleration_system [2018/02/21 15:06]
jiyuanz created
seminars:adaptive_load_balancer_for_distributed_fpga_acceleration_system [2018/02/21 15:08]
jiyuanz
Line 1: Line 1:
-Title: ​Adaptive Load Balancer for Distributed FPGA Acceleration System+====== ​Adaptive Load Balancer for Distributed FPGA Acceleration System ​====== 
 + 
 +Wednesday March 7, 2018\\ 
 +Location: HH D-level Conference Room\\ 
 +Time: 4:30PM\\ 
 + 
 + 
 + 
 +=====Abstract=====
  
-Abstract: ​ 
 FPGAs have been deployed in data center to accelerate cloud applications,​ for example, Microsoft Catapult FPGA system, where a FPGA is attached to each server and all the FPGAs are interconnected through Ethernet. When a server'​s host CPU offloads a job to the attached FPGA, it actually submits the job to a pool of FPGAs. Then the question is which FPGA should we choose to accelerate that job. This is a load balancing problem, because if some FPGAs are overloaded, the latency of the jobs would be high.  FPGAs have been deployed in data center to accelerate cloud applications,​ for example, Microsoft Catapult FPGA system, where a FPGA is attached to each server and all the FPGAs are interconnected through Ethernet. When a server'​s host CPU offloads a job to the attached FPGA, it actually submits the job to a pool of FPGAs. Then the question is which FPGA should we choose to accelerate that job. This is a load balancing problem, because if some FPGAs are overloaded, the latency of the jobs would be high. 
  
 We propose an adaptive load balancer that using the number of in-flight jobs to model the load of FPGAs. For our Bing search case study, we observed that our adaptive load balancer can (1)balance the load evenly and dynamically,​ (2) is insensitive to load traffic types, and (3)keep the system simple. Compared with the baseline hardware Round-robin load balancer, our adaptive load balancer can reduce Bing search 99.9% tail latency from 15% to 35%. This work was done when I was a summer intern at Microsoft Research Catapult Team in 2017.  We propose an adaptive load balancer that using the number of in-flight jobs to model the load of FPGAs. For our Bing search case study, we observed that our adaptive load balancer can (1)balance the load evenly and dynamically,​ (2) is insensitive to load traffic types, and (3)keep the system simple. Compared with the baseline hardware Round-robin load balancer, our adaptive load balancer can reduce Bing search 99.9% tail latency from 15% to 35%. This work was done when I was a summer intern at Microsoft Research Catapult Team in 2017. 
  
-Bio+=====Bio===== 
 Zhipeng Zhao is a fourth year Ph.D. student in the Department of Electrical and Computer Engineering at CMU advised by Prof. James Hoe. His research focuses on High-level Synthesis, distributed FPGA acceleration system and Network Function acceleration. Zhipeng Zhao is a fourth year Ph.D. student in the Department of Electrical and Computer Engineering at CMU advised by Prof. James Hoe. His research focuses on High-level Synthesis, distributed FPGA acceleration system and Network Function acceleration.