|An Efficient Method for Stream Semantics over RDMA||P. MacArthur and R. Russell||
Most network applications today are written to use TCP/IP via sockets. Remote Direct Memory Access (RDMA) is gaining popularity because its zero-copy, kernel-bypass features provide a high throughput, low latency reliable transport. Unlike TCP, which is a stream-oriented protocol, RDMA is a message-oriented protocol, and the OFA verbs library for writing RDMA application programs is more complex than the TCP sockets interface. UNH EXS is one of several libraries designed to give applications more convenient, high-level access to RDMA features. Recent work has shown that RDMA is viable both in the data center and over distance.
One potential bottleneck in libraries that use RDMA is the requirement to wait for message advertisements in order to send large zero-copy messages. By sending messages first to an internal, hidden buffer and copying the message later, latency can be reduced at the expense of higher CPU usage at the receiver. This paper presents a communication algorithm that has been implemented in the UNH EXS stream-oriented mode to allow dynamic switching between sending transfers directly to user memory and sending transfers indirectly via an internal, hidden buffer depending on the state of the sender and receiver. Based on preliminary results, we see that this algorithm performs well under a variety of application requirements.
|May 21, 2014||White Papers|
Epoch & Unix Timestamp Conversion Tools
|October 7, 2013||External Links|
|10Gig Link Fault Signaling||Eric Lynskey||
Short overview of how 10Gig link fault signaling works.
|September 5, 2013||Tutorials|
|Further Testing of TRILL (TRansparent Interconnection of Lots of Links)||Christina Dube||
The University of New Hampshire InterOperability Laboratory (UNH-IOL) hosted its second TRansparent Interconnection of Lots of Links (TRILL) Interoperability Test Event the week of May 20 - May 24, 2013 at its 32,000+ square-foot facility in Durham, New Hampshire. The test event brought together implementers of TRILL as well as test equipment manufacturers that support TRILL. The purpose of the test event was to gain a perspective on the current status of TRILL implementation and interoperability. Participants included Extreme Networks, Hewlett-Packard Networking, Huawei Technologies, and Ixia.
|May 31, 2016||White Papers|
|FQTSS Overview||William Gravelle||
FQTSS is defined in the IEEE standard 801.Q Clause 34; Forwarding and Queuing Enhancements for Time-Sensitive Streams. The term FQTSS is used to describe a set of tools which are used to forward and queue time-sensitive streams. Since AVB frames cannot be dropped, there must be a mechanism in place to forward AVB frames quickly and efficiently. This is where FQTSS (aka Qav) comes into play. This paper goes into the components that make up FQTSS.
|July 16, 2013||White Papers|
|Mellanox IB DDR Auto-negotiation Specification 1.0||
The specification for the Mellanox DDR Proprietary Auto-negotiation protocol.
|July 11, 2013||White Papers|
|Neighbor Discovery State Machine for the Reachability State||Andrew Gadzik, Thomas Peterson, Chris Cavanaugh||
This state machine flow chart covers a summary of the rules specified in RFC 4861 Sections 7.2 and 7.3.
|July 2, 2013||Tutorials|
|Testing TRILL (TRansparent Interconnect for Lots of Links)||Christina Dube||
The University of New Hampshire InterOperability Laboratory (UNH-IOL) hosted its second TRansparent Interconnect of Lots of Links (TRILL) Interoperability Test Event the week of November 26 Ã¢Â€Â“ December 1, 2012 at its 32,000+ square-foot facility in Durham, New Hampshire. The test event brought together implementers of TRILL as well as test equipment manufacturers that support TRILL. The purpose of the test event was to gain a perspective on the current status of TRILL implementation and interoperability. Participants included Hewlett-Packard Networking, Extreme Networks, Ixia and Spirent Communications.
|February 18, 2013||White Papers|
|A Performance Study to Guide RDMA Programming Decisions||Patrick MacArthur and Dr. Robert Russell||
This paper describes a performance study of Remote Direct Memory Access (RDMA) programming techniques. Its goal is to use these results as a guide for making "best practice" RDMA programming decisions. Infiniband RDMA is widely used in scientific high performance computing (HPC) clusters as a low-latency, high-bandwidth, reliable interconnect accessed via MPI. Recently it is gaining adherents outside scientific HPC as high-speed clusters appear in other application areas for which MPI is not suitable. RDMA enables user applications to move data
|January 18, 2013||White Papers|
The OPEN (One-Pair Ether-Net) Alliance is designed to encourage wide scale adoption of Ethernet-based, single pair unshielded networks as the standard in automotive applications. The jointly developed OPEN Alliance SIG (Special Interest Group) will address industry requirements for improving in-vehicle safety, comfort, and infotainment, while significantly reducing network complexity and cabling costs.
|January 2, 2013||External Links|