Chip multiprocessors (CMPs) combine increasingly many general-purpose processor cores on a single chip. These cores run several tasks with unpredictable communication needs, resulting in uncertain and often-changing traffic patterns. This unpredictability leads network-on-chip (NoC) designers to plan for the worst-case traffic patterns, and significantly over-provision link capacities. In this paper, we provide NoC designers with an alternative statistical approach. We first present the traffic-load distribution plots (T-Plots), illustrating how much capacity over-provisioning is needed to service 90%, 99%, or 100% of all traffic patterns. We prove that in the general case, plotting T-Plots is #P-complete, and therefore extremely complex. We then show how to determine the exact mean and variance of the traffic load on any edge, and use these to provide Gaussian-based models for the T-Plots, as well as guaranteed performance bounds. Finally, we use T-Plots to reduce the network power consumption by providing an efficient capacity allocation algorithm with predictable performance guarantees. © 2008 IEEE.
Statistical approach to NoC design / Cohen, I.; Rottenstreich, O.; Keslassy, I.. - (2008), pp. 171-180. (Intervento presentato al convegno 2nd IEEE International Symposium on Networks-on-Chip, NOCS 2008 tenutosi a Newcastle upon Tyne, gbr nel 2008) [10.1109/NOCS.2008.4492736].
Statistical approach to NoC design
Cohen I.;
2008
Abstract
Chip multiprocessors (CMPs) combine increasingly many general-purpose processor cores on a single chip. These cores run several tasks with unpredictable communication needs, resulting in uncertain and often-changing traffic patterns. This unpredictability leads network-on-chip (NoC) designers to plan for the worst-case traffic patterns, and significantly over-provision link capacities. In this paper, we provide NoC designers with an alternative statistical approach. We first present the traffic-load distribution plots (T-Plots), illustrating how much capacity over-provisioning is needed to service 90%, 99%, or 100% of all traffic patterns. We prove that in the general case, plotting T-Plots is #P-complete, and therefore extremely complex. We then show how to determine the exact mean and variance of the traffic load on any edge, and use these to provide Gaussian-based models for the T-Plots, as well as guaranteed performance bounds. Finally, we use T-Plots to reduce the network power consumption by providing an efficient capacity allocation algorithm with predictable performance guarantees. © 2008 IEEE.File | Dimensione | Formato | |
---|---|---|---|
NoC_Conf_08.pdf
accesso aperto
Tipologia:
2. Post-print / Author's Accepted Manuscript
Licenza:
Pubblico - Tutti i diritti riservati
Dimensione
252.78 kB
Formato
Adobe PDF
|
252.78 kB | Adobe PDF | Visualizza/Apri |
Cohen-StatisticalApproach.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
350.38 kB
Formato
Adobe PDF
|
350.38 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2873216