Stratus: cost-aware container scheduling in the public cloud

被引:60
作者
Chung, Andrew [1 ]
Park, Jun Woo [1 ]
Ganger, Gregory R. [1 ]
机构
[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA
来源
PROCEEDINGS OF THE 2018 ACM SYMPOSIUM ON CLOUD COMPUTING (SOCC '18) | 2018年
关键词
cloud computing; cluster scheduling; transient server; ENERGY;
D O I
10.1145/3267809.3267819
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Stratus is a new cluster scheduler specialized for orchestrating batch job execution on virtual clusters, dynamically allocated collections of virtual machine instances on public IaaS platforms. Unlike schedulers for conventional clusters, Stratus focuses primarily on dollar cost considerations, since public clouds provide effectively unlimited, highly heterogeneous resources allocated on demand. But, since resources are charged-for while allocated, Stratus aggressively packs tasks onto machines, guided by job runtime estimates, trying to make allocated resources be either mostly full (highly utilized) or empty (so they can be released to save money). Simulation experiments based on cluster workload traces from Google and TwoSigma show that Stratus reduces cost by 17-44% compared to state-of-the-art approaches to virtual cluster scheduling.
引用
收藏
页码:121 / 134
页数:14
相关论文
共 57 条
[1]  
AMVROSIADIS G., 2018, 2018 USENIX ANN TECH
[2]  
[Anonymous], 2011, P INT C HIGH PERF CO
[3]  
[Anonymous], 2018, HOW SPOT FLEET WORKS
[4]  
[Anonymous], 2010, 2010 IEEE 3 INT C CL
[5]  
[Anonymous], P 4 ANN S CLOUD COMP, DOI [10.1145/2523616.2523633, DOI 10.1145/2523616.2523633]
[6]  
[Anonymous], 2018, AWS EC2 SPOT INSTANC
[7]  
[Anonymous], 2018, AZURE VIRTUAL MACHIN
[8]  
[Anonymous], 2018, AMAZON ECS TASK PLAC
[9]  
[Anonymous], 2004, OSDI 04
[10]  
[Anonymous], 2018, POWERING YOUR AMAZON