The Experts below are selected from a list of 2289 Experts worldwide ranked by ideXlab platform
Masahito Shiba - One of the best experts on this subject based on the ideXlab platform.
-
GCCE - An Examination Method of Mergeable Memory Page Distribution for Memory Deduplication
2019 IEEE 8th Global Conference on Consumer Electronics (GCCE), 2019Co-Authors: Masahito ShibaAbstract:When multiple virtual machines run in a single Physical machine, the same libraries and applications are often used and these virtual machines have Memory Pages of the same contents. Memory deduplication, which merge such Memory Pages that have the same contents into one Physical Memory Page, can improve the efficiency of the Memory usage. However, the cost of finding mergeable Pages is significantly expensive. This paper describes a method to investigate how mergeable Memory Pages are distributed in address spaces. The method classifies the Pages by state of consecutive Pages and measures the proportion of the Pages in each state. This paper also describes that the actual Page distribution measured by the proposed method on environment that virtual machines run. It shows that measuring the distribution is useful for estimating the probability of merging Pages for Memory deduplication.
Bob B. Aglietti - One of the best experts on this subject based on the ideXlab platform.
-
SC - Dynamic Page Placement to Improve Locality in CC-NUMA Multiprocessors for TPC-C
Proceedings of the 2001 ACM IEEE conference on Supercomputing (CDROM) - Supercomputing '01, 2001Co-Authors: Kenneth Mark Wilson, Bob B. AgliettiAbstract:The use of CC-NUMA multiprocessors complicates the placement of Physical Memory Pages. Memory closest to a processor provides the best access time, but optimal Memory Page placement is a difficult problem with process movement, multiple processes requiring access to the same Physical Memory Page, and application behavior changing over execution time. We use dynamic Page placement to move Memory Pages where needed for the database benchmark TPC-C executing on a four node CC-NUMA multiprocessor. Dynamic Page placement achieves local Memory accesses up to 73% of the time instead of the static Page placement results of 34% locality achieved with first touch and 25% with round robin. This can result in a 17% improvement in performance.
Kenneth Mark Wilson - One of the best experts on this subject based on the ideXlab platform.
-
SC - Dynamic Page Placement to Improve Locality in CC-NUMA Multiprocessors for TPC-C
Proceedings of the 2001 ACM IEEE conference on Supercomputing (CDROM) - Supercomputing '01, 2001Co-Authors: Kenneth Mark Wilson, Bob B. AgliettiAbstract:The use of CC-NUMA multiprocessors complicates the placement of Physical Memory Pages. Memory closest to a processor provides the best access time, but optimal Memory Page placement is a difficult problem with process movement, multiple processes requiring access to the same Physical Memory Page, and application behavior changing over execution time. We use dynamic Page placement to move Memory Pages where needed for the database benchmark TPC-C executing on a four node CC-NUMA multiprocessor. Dynamic Page placement achieves local Memory accesses up to 73% of the time instead of the static Page placement results of 34% locality achieved with first touch and 25% with round robin. This can result in a 17% improvement in performance.
Tzi-cker Chiueh - One of the best experts on this subject based on the ideXlab platform.
-
ICPADS - Hypervisor Support for Efficient Memory De-duplication
2011 IEEE 17th International Conference on Parallel and Distributed Systems, 2011Co-Authors: Yingshiuan Pan, Jui-hao Chiang, Po-jui Tsao, Ming-fen Lin, Tzi-cker ChiuehAbstract:Memory de-duplication removes the Memory state redundancy among virtual machines that run on the same Physical machine by identifying common Memory Pages shared by these virtual machines and storing only one copy for each of common Memory Pages. A standard approach to identifying common Memory Pages is to hash the content of each Memory Page, and compare the resulting hash values. In a virtualized server, only the hyper visor is in a position to compute the hash value of every Physical Memory Page on the server, but the Memory de-duplication engine is best implemented outside the hyper visor for flexibility and simplicity reasons. A key design issue in the Memory de-duplication engine is to minimize the performance impact of these hashing computations on the running VMs. To reduce this impact, Memory Page hashing should be performed with low overhead and when the CPU is idle. This paper describes why existing hyper visors do not provide adequate support for our Memory de-duplication engine, how a new primitive called deferrable aggregate hyper call (DAH) fills the need, and what the resulting performance improvement is.
Yingshiuan Pan - One of the best experts on this subject based on the ideXlab platform.
-
ICPADS - Hypervisor Support for Efficient Memory De-duplication
2011 IEEE 17th International Conference on Parallel and Distributed Systems, 2011Co-Authors: Yingshiuan Pan, Jui-hao Chiang, Po-jui Tsao, Ming-fen Lin, Tzi-cker ChiuehAbstract:Memory de-duplication removes the Memory state redundancy among virtual machines that run on the same Physical machine by identifying common Memory Pages shared by these virtual machines and storing only one copy for each of common Memory Pages. A standard approach to identifying common Memory Pages is to hash the content of each Memory Page, and compare the resulting hash values. In a virtualized server, only the hyper visor is in a position to compute the hash value of every Physical Memory Page on the server, but the Memory de-duplication engine is best implemented outside the hyper visor for flexibility and simplicity reasons. A key design issue in the Memory de-duplication engine is to minimize the performance impact of these hashing computations on the running VMs. To reduce this impact, Memory Page hashing should be performed with low overhead and when the CPU is idle. This paper describes why existing hyper visors do not provide adequate support for our Memory de-duplication engine, how a new primitive called deferrable aggregate hyper call (DAH) fills the need, and what the resulting performance improvement is.