The SP1 high-performance switch

Stunkel, Craig B.; Shea, Dennis G.; Grice, Don; Hochschild, Peter; Tsao, M.

doi:10.1109/shpcc.1994.296638

Cited by 68 publications

(24 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2a illustrates source routing where a packet is routed through nodes A, B, and C. The route corresponds to East, North, and North, which is reflected in the routing flits in the packet. Thanks to its speed and simplicity, source routing has been widely used in many router designs including the IBM SP1 [29] and the Avici TSR [6], although it leads to longer packets containing routing flits as compared to the case where the route is computed for each hop.…”

Section: Router Architecture For Application-awarementioning

confidence: 99%

See 1 more Smart Citation

Optimal and Heuristic Application-Aware Oblivious Routing

Kinsy

Cho

Shim

et al. 2013

IEEE Trans. Comput.

View full text Add to dashboard Cite

Abstract-Conventional oblivious routing algorithms do not take into account resource requirements (e.g., bandwidth, latency) of various flows in a given application. As they are not aware of flow demands that are specific to the application, network resources can be poorly utilized and cause serious local congestion. Also, flows, or packets, may share virtual channels in an undetermined way; the effects of head-of-line blocking may result in throughput degradation. In this paper, we present a framework for application-aware routing that assures deadlock freedom under one or more virtual channels by forcing routes to conform to an acyclic channel dependence graph. In addition, we present methods to statically and efficiently allocate virtual channels to flows or packets, under oblivious routing, when there are two or more virtual channels per link. Using the application-aware routing framework, we develop and evaluate a bandwidth-sensitive oblivious routing scheme that statically determines routes considering an application's communication characteristics. Given bandwidth estimates for flows, we present a mixed integer-linear programming (MILP) approach and a heuristic approach for producing deadlock-free routes that minimize maximum channel load. Our framework can be used to produce application-aware routes that target the minimization of latency, number of flows through a link, bandwidth, or any combination thereof. Our results show that it is possible to achieve better performance than traditional deterministic and oblivious routing schemes on popular synthetic benchmarks using our bandwidth-sensitive approach. We also show that, when oblivious routing is used and there are more flows than virtual channels per link, the static assignment of virtual channels to flows can help mitigate the effects of head-ofline blocking, which may impede packets that are dynamically competing for virtual channels. We experimentally explore the performance tradeoffs of static and dynamic virtual channel allocation on bandwidth-sensitive and traditional oblivious routing methods.

show abstract

Section: Router Architecture For Application-awarementioning

confidence: 99%

“…Both routing methods, namely source routing and nodetable routing, are widely known and have been implemented in multiple routers [29], [6], [11]. In other words, the proposed routing approach can be realized with standard routing hardware without new specialized mechanisms.…”

Section: Router Architecture For Application-awarementioning

confidence: 99%

Optimal and Heuristic Application-Aware Oblivious Routing

Kinsy

Cho

Shim

et al. 2013

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…Both routing methods are widely known and have been implemented in multiple routers [39,38,7,13]. In other words, the proposed routing approach can be realized with standard routing hardware without new specialized mechanisms.…”

Section: Programmable Routingmentioning

confidence: 99%

Application-aware deadlock-free oblivious routing

Kinsy

Cho

Wen

et al. 2009

Proceedings of the 36th Annual International Symposium on Computer Architecture

View full text Add to dashboard Cite

Conventional oblivious routing algorithms are either not application-aware or assume that each flow has its own private channel to ensure deadlock avoidance. We present a framework for application-aware routing that assures deadlock-freedom under one or more channels by forcing routes to conform to an acyclic channel dependence graph. Arbitrary minimal routes can be made deadlock-free through appropriate static channel allocation when two or more channels are available. Given bandwidth estimates for flows, we present a mixed integer-linear programming (MILP) approach and a heuristic approach for producing deadlock-free routes that minimize maximum channel load. The heuristic algorithm is calibrated using the MILP algorithm and evaluated on a number of benchmarks through detailed network simulation. Our framework can be used to produce application-aware routes that target the minimization of latency, number of flows through a link, bandwidth, or any combination thereof.

show abstract

“…Many MIN families were studied and effectively used to build parallel computers [11,1,34]. Delta MINs form a subgroup of a bigger MIN family called Banyan MINs.…”

Section: Introductionmentioning

confidence: 99%

A universal performance factor for multi-criteria evaluation of multistage interconnection networks

Aljundi

Dekeyser

Kechadi

et al. 2006

Future Generation Computer Systems

View full text Add to dashboard Cite

The choice of an interconnection network for a parallel computer depends on a large number of performance factors which are very often application dependent. We propose a performance evaluation and a comparison methodology. This methodology is applied to a recently introduced class of interconnection networks (Multistage Chordal Ring Based multistage interconnection network). These networks are compared to the well known Omega network of comparable architectural characteristics. The methodology is expected to serve in the evaluation of the use of multistage interconnection networks as an intercommunication medium in today's multi-processor systems.

show abstract

The SP1 high-performance switch

Cited by 68 publications

References 0 publications

Optimal and Heuristic Application-Aware Oblivious Routing

Optimal and Heuristic Application-Aware Oblivious Routing

Application-aware deadlock-free oblivious routing

A universal performance factor for multi-criteria evaluation of multistage interconnection networks

Contact Info

Product

Resources

About