1. August 22, 2005Page 1 of (#) Datacenter Fabric Workshop Open MPI Overview and Current Status Tim Woodall - LANL Galen Shipman - LANL/UNM

2. Page 2 Overview Point-to-Point Architecture OpenIB –Implementation –Results Future Work

3. Page 3 Point-to-Point Architecture Component Architecture: – “Plug-ins” for different capabilities (e.g. different networks) –Tunable run-time parameters Three component frameworks: –Point-to-point messaging layer (PML) implements MPI semantics –Byte Transfer Layer (BTL) abstracts network interfaces –Memory Pool (mpool) provides for memory management/registration

4. Page 4 PML Framework Single PML manages multiple BTL modules –Maintains set of BTLs on a per-peer basis –Message fragmentation and scheduling Implements MPI semantics −Synchronous / buffered / ready / normal sends −Persistent requests / Request completion Eager/Rendezvous protocol −Eager send of short messages −Configurable threshold (short vs. long) −Multiple long protocols

5. Page 5 PML Protocols Send / receive pipeline to / from pre-registered buffers (non-contiguous data) MPI_Alloc_mem support –Red/black tree of memory registrations –BTL associated with registration is used by scheduler –Xfer of contiguous data with 1 RDMA (after match) “Leave pinned” run-time parameter –Registration on first-use –MRU cache (configurable size) of registrations –Bandwidth equivalent to pre-registered buffers (MPI_Alloc_mem)

6. Page 6 PML Protocols (Continued) Dynamic memory registration/deregistration –Fragment message and build pipeline of RDMA requests –Overlap [de-]registration with RDMA –Bandwidth 97% of pre-registered memory at large message sizes (8Mbytes) –Performance impacted by bus type/bandwidth

7. Page 7 BTL Framework MPI agnostic Provides simple API to upper layers –Tagged send/receive primitives –One-sided put/get operations Access to data type engine for zero copy data transfer BTL modules natively support commodity networks: –Current (self, shared memory, myrinet GM/MX, Infiniband mvapi/OpenIB, Portals, TCP) –Planned (LAPI, Quadrics Elan4)

8. Page 8 OpenIB BTL BTL module initialization Resources allocation Connection management Small message Xfer Large message Xfer OpenIB Issues Future Work

9. Page 9 BTL module initialization A separate BTL module is initialized for each port on each HCA The PML schedules across these BTL modules just as any other interconnect When multiple BTL modules exist peers establish QP connections by matching subnets

10. Page 10 Resource Allocation

11. Page 11 SRQ Scalability 5105126481024 48256648512 24128648256 1264 8128 SRQ- Mbytes K*RQ per QP- Mbytes #postedFrag size- Kbytes Nodes K- multiplier based on number of nodes

12. Page 12 Connection management Addressing information is exchanged dynamically via an OOB channel –This greatly improves scalability but at the cost of increased first message latency –Connections are established with peers in the same subnet (local subnet routing only)

13. Page 13 Small Message Xfer –Maintain list of pre-registered fragments for send and recv –List grows dynamically in chunks (more efficient to register) –Small messages are copied to/from pre- registered buffers –Recv descriptors are posted as needed based on min/max thresholds

14. Page 14 Small Message Performance Average Latency OpenMPI - OpenIB - *optimized5.13 usec OpenMPI - OpenIB - *defaults5.43 usec OpenMPI - Mvapi - *optimized5.64 usec OpenMPI - Mvapi - *defaults5.94 usec Mvapich - Mvapi (rdma/mem poll)4.19 usec Mvapich - Mvapi (send/recv)6.51 usec * Send/Recv based protocol

15. Page 15 Large Message Xfer RDMA Write and RDMA Read are both supported RDMA Read provides better performance than RDMA Write - control messages are reduced RDMA pipeline protocol performance highly dependent on I/O Bus performance

16. Page 16 Results OpenMPI/OpenIB - All

17. Page 17 Results OpenMPI/OpenIB - All - Log

18. Page 18 Results OpenMPI/OpenIB - Eager limit

19. Page 19 Results Combined Results

20. Page 20 Results Combined Results - Log

21. Page 21 OpenIB Opportunities –User level notification of VM activity Caching of memory registrations can be dangerous Need the ability to detect VM changes that effect memory registrations (such as sbrk and munmap) –Reliable Multicast for collectives –SRQ performance, 2/10 usec penalty, but who’s counting?

22. Page 22 Future Work Small message RDMA (using working set of peers) - optional Dynamic connection management using Unreliable Datagrams Dynamic connection teardown - optional

23. Page 23 Source Code Access Subversion repository Download client from: –http://subversion.tigris.org/ –v1.2.1 or later Check out with: –svn co http://svn.open-mpi.org/svn/ompi/trunk ompi –Anonymous, read-only access

24. Page 24 Questions? Tim Woodall Email: twoodall@lanl.gov Phone: 505-665-5224 Galen Shipman Email: gshipman@lanl.gov

25. Page 25 Hardware Specs Dual Intel Xeon 3.2 GHz –1024 KB Cache 2 Gbytes memory Bus: Intel Corp. E7525/E7520/E7320 PCI Express Mellanox Technologies MT25208 InfiniHost III Ex 288 Port Voltaire switch