Filters








113 Hits in 7.7 sec

Communicating Efficiently on Cluster-Based Remote Direct Memory Access (RDMA) over InfiniBand Protocol

Masoud Hemmatpour, Bartolomeo Montrucchio, Maurizio Rebaudengo
2018 Applied Sciences  
Designing distributed applications over such protocols requires a fundamental rethinking in communication components in comparison with traditional protocols (i.e., TCP/IP).  ...  Distributed systems are commonly built under the assumption that the network is the primary bottleneck, however this assumption no longer holds by emerging high-performance RDMA enabled protocols in datacenters  ...  We would like to say thanks to Dotan Barak for providing helpful advice on the RDMA implementation on InfiniBand adapters, and Aleksandar Dragojevic for discussing fundamental issues on RDMA.  ... 
doi:10.3390/app8112034 fatcat:5ptfifhqmnhmtojc7yax4r6ewi

Reliable multicast using remote direct memory access (RDMA) over a passive optical cross-connect fabric enhanced with wavelength division multiplexing (WDM)

Kin-Wai Leong, Zhilong Li, Yunqu Leon Liu
2019 APSIPA Transactions on Signal and Information Processing  
In this paper, we introduce a layer-4 (L4) transport based on remote direct memory access (RDMA) datagram to achieve reliable multicast over a shared optical medium.  ...  Furthermore, with this reliable multicast transport, end-to-end latencies of eight microseconds or less (< 8us) have been routinely achieved using an enhanced software RDMA implementation on a variety  ...  In Section III, we discuss a low-loss and low latency implementation by combining ODBSS with the use of remote direct memory access (RDMA) datagram.  ... 
doi:10.1017/atsip.2019.17 fatcat:scg4jiznkjfmfehildtga5ln2y

Minimizing the Hidden Cost of RDMA

Philip Werner Frey, Gustavo Alonso
2009 2009 29th IEEE International Conference on Distributed Computing Systems  
Remote Direct Memory Access (RDMA) is a mechanism whereby data is moved directly between the application memory of the local and remote computer.  ...  With the advent of hardware implementations of RDMA over Ethernet (iWARP), its advantages have become even more obvious.  ...  Introduction Remote Direct Memory Access (RDMA) is a mechanism whereby data is placed directly in the application memory of a remote computer [1] .  ... 
doi:10.1109/icdcs.2009.32 dblp:conf/icdcs/FreyA09 fatcat:6czonsoufzcajh4nizafjio2ku

A study of iSCSI extensions for RDMA (iSER)

Mallikarjun Chadalapaka, Hemal Shah, Uri Elzur, Patricia Thaler, Michael Ko
2003 Proceedings of the ACM SIGCOMM workshop on Network-I/O convergence experience, lessons, implications - NICELI '03  
The iWARP protocol suite provides Remote Direct Memory Access (RDMA) semantics over TCP/IP networks and enables efficient memory-to-memory data transfers over an IP fabric.  ...  This paper studies the design process of iSCSI Extensions for RDMA (iSER), a protocol that maps the iSCSI protocol over the iWARP protocol suite.  ...  Acknowledgments This protocol was developed by a design team that, in addition to the authors, includes the members of the storage group in the RDMA Consortium.  ... 
doi:10.1145/944747.944754 fatcat:mplcybqyi5cvxmvbgm4sfc2ghq

A study of iSCSI extensions for RDMA (iSER)

Mallikarjun Chadalapaka, Hemal Shah, Uri Elzur, Patricia Thaler, Michael Ko
2003 Proceedings of the ACM SIGCOMM workshop on Network-I/O convergence experience, lessons, implications - NICELI '03  
The iWARP protocol suite provides Remote Direct Memory Access (RDMA) semantics over TCP/IP networks and enables efficient memory-to-memory data transfers over an IP fabric.  ...  This paper studies the design process of iSCSI Extensions for RDMA (iSER), a protocol that maps the iSCSI protocol over the iWARP protocol suite.  ...  Acknowledgments This protocol was developed by a design team that, in addition to the authors, includes the members of the storage group in the RDMA Consortium.  ... 
doi:10.1145/944752.944754 fatcat:67le23mn3jd63osjjjvrbw5cvy

Rethinking database high availability with RDMA networks

Erfan Zamanian, Xiangyao Yu, Michael Stonebraker, Tim Kraska
2019 Proceedings of the VLDB Endowment  
Active-Memory maintains high availability and correctness in the presence of failures through an efficient RDMA-based undologging scheme.  ...  We present Active-Memory Replication, a new high availability scheme that efficiently leverages RDMA to completely eliminate the processing redundancy in replication.  ...  The latest remote direct memory access (RDMA) based networks, for example, achieve a bandwidth similar to that of main memory, while having only a factor of 10× higher latency.  ... 
doi:10.14778/3342263.3342639 fatcat:asul3szpqba5daj6eum2d2xvqu

High-Performance Design of HBase with RDMA over InfiniBand

Jian Huang, Xiangyong Ouyang, Jithin Jose, Md. Wasi-ur-Rahman, Hao Wang, Miao Luo, Hari Subramoni, Chet Murthy, Dhabaleswar K. Panda
2012 2012 IEEE 26th International Parallel and Distributed Processing Symposium  
These interconnects provide advanced network features, such as Remote Direct Memory Access (RDMA), to achieve high throughput and low latency along with low CPU utilization.  ...  RDMA follows memory-block semantics, which can be adopted efficiently to satisfy the object transmission primitives used in HBase.  ...  One of the main features of InfiniBand is Remote Direct Memory Access (RDMA).  ... 
doi:10.1109/ipdps.2012.74 dblp:conf/ipps/HuangOJWWLSMP12 fatcat:owbotqyffndlrebgbrscn5kepm

The End of a Myth: Distributed Transactions Can Scale [article]

Erfan Zamanian, Carsten Binnig, Tim Kraska, Tim Harris
2016 arXiv   pre-print
Our experiments with the TPC-C benchmark show that our system scales linearly to over 6.5 million new-order (14.5 million total) distributed transactions per second on 56 machines.  ...  present the design of our novel scalable database system NAM-DB and show that distributed transactions with the very common Snapshot Isolation guarantee can indeed scale using the next generation of RDMA-enabled  ...  InfiniBand offers two network communication stacks: IP over InfiniBand (IPoIB) and remote direct memory access (RDMA).  ... 
arXiv:1607.00655v2 fatcat:4aktawyrczchfg65ojcabzcfhi

The end of a myth

Erfan Zamanian, Carsten Binnig, Tim Harris, Tim Kraska
2017 Proceedings of the VLDB Endowment  
Our experiments with the TPC-C benchmark show that our system scales linearly to over 6.5 million new-order (14.5 million total) distributed transactions per second on 56 machines.  ...  present the design of our novel scalable database system NAM-DB and show that distributed transactions with the very common Snapshot Isolation guarantee can indeed scale using the next generation of RDMA-enabled  ...  System Overview InfiniBand offers two network communication stacks: IP over InfiniBand (IPoIB) and remote direct memory access (RDMA).  ... 
doi:10.14778/3055330.3055335 fatcat:v3yfkdn2fbf5nhjvpq5i6znf44

Scalable Work Stealing of Native Threads on an x86-64 Infiniband Cluster

Shigeki Akiyama, Kenjiro Taura
2016 Journal of Information Processing  
Furthermore, this huge virtual memory consumption makes it impossible to implement one-sided work stealing with Remote Direct Memory Access (RDMA) operations.  ...  Task parallelism on large-scale distributed memory environments is still a challenging problem. The focuses of our work are flexibility of task model and scalability of inter-node load balancing.  ...  Direct Memory Access (RDMA).  ... 
doi:10.2197/ipsjjip.24.583 fatcat:lrsku2xz4feadb74xmlwzjqvrq

High Performance Data Transfer in Grid Environment Using GridFTP over InfiniBand

Hari Subramoni, Ping Lai, Raj Kettimuthu, Dhabaleswar K. Panda
2010 2010 10th IEEE/ACM International Conference on Cluster, Cloud and Grid Computing  
On the other hand, modern interconnects such as InfiniBand have introduced many advanced communication features, like zero-copy protocol and RDMA operations which can greatly improve the communication  ...  In this paper, we take on the challenge of combining the ease of use of the Globus XIO framework and the high performance achieved through InfiniBand communication, thereby natively supporting GridFTP over  ...  Remote Direct Memory Access (RDMA)  ... 
doi:10.1109/ccgrid.2010.115 dblp:conf/ccgrid/SubramoniLKP10 fatcat:5ebdlsdzvzdtvaeatdnlza3iti

Network Processing on an SPE Core in Cell Broadband Engine

Yuji Kawamura, Takeshi Yamazaki, Hiroshi Kyusojin, Tatsuya Ishiwata, Kazuyoshi Horie
2008 2008 16th IEEE Symposium on High Performance Interconnects  
processing and memory copies.  ...  Our work shows that the use of the SPE instead of the PPE for network processing can help resolve network performance problems that can arise from handling a highspeed NIC, including the costs of protocol  ...  Direct Memory Access protocol API (RDMA API) Figure 7 7 shows the results of TCP loopback communications.  ... 
doi:10.1109/hoti.2008.16 dblp:conf/hoti/KawamuraYKIH08 fatcat:7wkyw4b2mnhjlj5yqptmn7xjhe

"Implementation Notes For The Storage And Data Access Infrastructure", Exanest Project Deliverable D4.3

ExaNeSt Consortium
2017 Zenodo  
Two main technologies for hardware-assisted virtualisation are considered: (i) use of RDMA capabilities for accessing remote memory from within a virtual machine, and (ii) HPC API remoting to improve the  ...  kernel related overheads; and (ii) enhancements to the access path for memory-mapped file access by, amongst others, making use of NVM devices.  ...  Two main technologies for hardware-assisted virtualisation are considered: i) use of RDMA capabilities for accessing remote memory from within a virtual machine, and ii) HPC API remoting to improve the  ... 
doi:10.5281/zenodo.802198 fatcat:x4ohfmywmjdjjlxogciiq27o7i

Hybrid Programming Using OpenSHMEM and OpenACC [chapter]

Matthew Baker, Swaroop Pophale, Jean-Charles Vasnier, Haoqiang Jin, Oscar Hernandez
2014 Lecture Notes in Computer Science  
Previous efforts looking into hybrid models have primarily focused on using OpenMP directives (for shared memory programming) with MPI (for inter-node programming on a cluster), using OpenMP to spawn threads  ...  In this paper we explore the use of OpenACC directives to program GPUs and the use of OpenSHMEM, a PGAS library for onesided communication between nodes.  ...  memory access (RDMA).  ... 
doi:10.1007/978-3-319-05215-1_6 fatcat:fvma7qrokndwveorev7pq7jcqu

Addressing System Bottlenecks [chapter]

Alexander Supalov, Andrey Semin, Michael Klemm, Christopher Dahnken
2014 Optimizing HPC Applications with Intel® Cluster Tools  
and high-speed fabrics, such as InfiniBand, for low-latency and high-bandwidth remote direct memory access (RDMA) traffic of the user applications.  ...  Note that remote direct memory access to a different node takes approximately 15 times longer than access to the memory of the remote socket, and it takes 27 times longer than access to the local memory  ... 
doi:10.1007/978-1-4302-6497-2_4 fatcat:lx5puw5lxfgldhduymxvn5rwfy
« Previous Showing results 1 — 15 out of 113 results