PPT-Data Locality Optimizations

for NUMA Systems Zoltán Majó Thomas R Gross Department of Computer Science ETH Zurich Switzerland 1 A Library for Portable and Composable NUMA systems Key to good performance data locality. Optimization. Techniques. . Presented by . Preethi Rajaram. CSS 548 Introduction to Compilers . Professor Carol Zander. Fall 2012 . Why?. Processor Speed . -. increasing at a faster rate than the memory speed. ExaScale. . Challenges and Potential Solutions. Bill Dally. Chief Scientist, . NVIDIA. Bell Professor of Engineering, Stanford University. Two Key Challenges. Programmability. Writing an efficient parallel program is hard. A General Analytical Study. Yongkun. Li, . Patrick P. C. Lee. , John C. S. . Lui. , . Yinlong. Xu. The Chinese . University . of Hong . Kong. University of Science and Technology of China. . 1. SSD Storage. Zhenhua . Guo. , Geoffrey Fox, Mo Zhou. Outline. Introduction. Data Locality and Fairness. Experiments. Conclusions. MapReduce Execution Overview. 3. Google File System. Read input data. Data locality. Zhenhua . Guo. , Geoffrey Fox, Mo Zhou. Outline. Introduction. Analysis of Data Locality. Optimality of Data Locality. Experiments. Conclusions. MapReduce Execution Overview. 3. Google File System. Read input data. •Data Locality –Temporal: if data item needed now, it is likely to be needed again in near future –Spatial: if data item needed now, nearby data likely to be needed in near future • in . Dyninst. Andrew . Bernat. , Matthew Legendre. Instrumentation is Complicated. User perspective: . “Insert some new code here, here, and here.”. Dyninst’s. perspective:. Relocation. – Move code to make space for instrumentation. Zhenhua . Guo. , Geoffrey Fox, Mo Zhou. Outline. Introduction. Analysis of Data Locality. Optimality of Data Locality. Experiments. Conclusions. MapReduce Execution Overview. 3. Google File System. Read input data. Research at UNT. . . Krishna . Kavi. Professor. Director of NSF Industry/University Cooperative Center . for Net-Centric Software and Systems (Net-Centric IUCRC). Computer Science and Engineering. The University of North Texas. George Kurian. 1. , . Srinivas. . Devadas. 1. , Omer . Khan. 2. , . 1 . Massachusetts Institute of Technology. 2 . University of Connecticut, . Storrs. 1. The Problem. Future multicore processors will have 100s of cores. Zhenhua . Guo. , Geoffrey Fox, Mo Zhou. Outline. Introduction. Data Locality and Fairness. Experiments. Conclusions. MapReduce Execution Overview. 3. Google File System. Read input data. Data locality. Bulk-synchronous Programming Models on. CPU Architectures. Hee-Seok. Kim. , . Izzat. El Hajj, John Stratton,. Steven . Lumetta. and . Wen-mei. . Hwu. CPU. (Intel, AMD, PowerPC). GPU. (NVIDIA, AMD, Imagination, ARM). Balazs Voneki. CERN/EP/LHCb Online group. TIPP2017, Beijing. 23.05.2017. LHCb Upgrade. Early results of RDMA optimizations on top of 100 Gbps . Ethernet, TIPP2017, Beijing, 23.05.2017 . – Balazs Voneki. Need to get Processing closer to storage. Need to get tasks close to data. Rack locality: . Hadoop. Kill task and if a local slot is available: Quincy. Why?. Network is bad: gives horrible performance.