Portable SHMEMCache: A High-Performance Key-Value Store on OpenSHMEM and MPI
- Florida State Univ., Tallahassee, FL (United States)
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
The integration of Big Data frameworks and HPC capabilities has drawn enormous interests in recent years. SHMEMCache is a distributed key-value store built on the OpenSHMEM global address space. It has solved several practical issues in leveraging OpenSHMEM’s one-sided operations for a distributed key-value store and providing efficient key-value operations on both commodity machines and supercomputers. However, being based solely on OpenSHMEM, SHMEMCache cannot leverage one-sided operations from a variety of software packages. This results in several limitations for SHMEMCache. First, we cannot make SHMEMCache available to a wider range of platforms. Second, an opportunity for potential performance improvement is missed. Third, there is a lack of deep understanding about how different one-sided operations can fit in with SHMEMCache and other distributed key-values in general. For example, the one-sided operations in OpenSHMEM and MPI have many differences in their interfaces, memory semantics and synchronization methods, all of which can have distinct implications and also increase the complexity in supporting both OpenSHMEM and MPI for SHMEMCache. Therefore, we have taken on an effort on leveraging different one-sided operations for SHMEMCache and proposed a design of portable SHMEMCache. Based on this new framework, we have supported both OpenSHMEM and MPI for SHMEMCache. We have also conducted an extensive set of experiments to compare the performance of the two versions on both commodity machines and the Titan supercomputer.
- Research Organization:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF); UT-Battelle LLC/ORNL, Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1567475
- Journal Information:
- OPENSHMEM AND RELATED TECHNOLOGIES: BIG COMPUTE AND BIG DATA CONVERGENCE, OPENSHMEM 2017, Vol. 10679; Conference: OpenSHMEM 2017: OpenSHMEM and Related Technologies. Big Compute and Big Data Convergence, Annapolis, Maryland, August 7-9, 2017.
- Country of Publication:
- United States
- Language:
- English
Providing a cloud network infrastructure on a supercomputer
|
conference | January 2010 |
Introducing OpenSHMEM: SHMEM for the PGAS community
|
conference | January 2010 |
Fast and general distributed transactions using RDMA and HTM
|
conference | January 2016 |
Benchmarking cloud serving systems with YCSB
|
conference | January 2010 |
An implementation and evaluation of the MPI 3.0 one-sided communication interface: MPI 3.0 RMA IMPLEMENTATION AND EVALUATION
|
journal | January 2016 |
No compromises: distributed transactions with consistency, availability, and performance
|
conference | January 2015 |
High-Performance Key-Value Store on OpenSHMEM
|
conference | May 2017 |
Implementing OpenSHMEM Using MPI-3 One-Sided Communication
|
book | January 2014 |
High-Performance Design of HBase with RDMA over InfiniBand
|
conference | May 2012 |
High performance MPI-2 one-sided communication over infinilband
|
conference | January 2004 |
Scalable Memcached Design for InfiniBand Clusters Using Hybrid Transports
|
conference | May 2012 |
Memcached Design on High Performance RDMA Capable Interconnects
|
conference | September 2011 |
High Performance RDMA-Based MPI Implementation over InfiniBand
|
journal | June 2004 |
Advances, Applications and Performance of the Global Arrays Shared Memory Programming Toolkit
|
journal | May 2006 |
Co-array Fortran for parallel programming
|
journal | August 1998 |
High-Performance Hybrid Key-Value Store on Modern Clusters with RDMA Interconnects and SSDs: Non-blocking Extensions, Designs, and Benefits
|
conference | May 2016 |
Infiniband scalability in Open MPI
|
conference | January 2006 |
Hadoop acceleration through network levitated merge
|
conference | January 2011 |
HydraDB: a resilient RDMA-driven key-value middleware for in-memory cluster computing
|
conference | January 2015 |
Similar Records
OpenSHMEM over MPI-3
Accelerating k-NN Algorithm with Hybrid MPI and OpenSHMEM