Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Design and Analysis of On-Chip Communication for Network-on-Chip Platforms
KTH, Skolan för informations- och kommunikationsteknik (ICT), Elektronik- och datorsystem, ECS.ORCID-id: 0000-0003-0061-3475
2007 (engelsk)Doktoravhandling, med artikler (Annet vitenskapelig)
Abstract [en]

Due to the interplay between increasing chip capacity and complex applications, System-on-Chip (SoC) development is confronted by severe challenges, such as managing deep submicron effects, scaling communication architectures and bridging the productivity gap. Network-on-Chip (NoC) has been a rapidly developed concept in recent years to tackle the crisis with focus on network-based communication. NoC problems spread in the whole SoC spectrum ranging from specification, design, implementation to validation, from design methodology to tool support. In the thesis, we formulate and address problems in three key NoC areas, namely, on-chip network architectures, NoC network performance analysis, and NoC communication refinement.

Quality and cost are major constraints for micro-electronic products, particularly, in high-volume application domains. We have developed a number of techniques to facilitate the design of systems with low area, high and predictable performance. From flit admission and ejection perspective, we investigate the area optimization for a classical wormhole architecture. The proposals are simple but effective. Not only offering unicast services, on-chip networks should also provide effective support for multicast. We suggest a connection-oriented multicasting protocol which can dynamically establish multicast groups with quality-of-service awareness. Based on the concept of a logical network, we develop theorems to guide the construction of contention-free virtual circuits, and employ a back-tracking algorithm to systematically search for feasible solutions.

Network performance analysis plays a central role in the design of NoC communication architectures. Within a layered NoC simulation framework, we develop and integrate traffic generation methods in order to simulate network performance and evaluate network architectures. Using these methods, traffic patterns may be adjusted with locality parameters and be configured per pair of tasks. We propose also an algorithm-based analysis method to estimate whether a wormhole-switched network can satisfy the timing constraints of real-time messages. This method is built on traffic assumptions and based on a contention tree model that captures direct and indirect network contentions and concurrent link usage.

In addition to NoC platform design, application design targeting such a platform is an open issue. Following the trends in SoC design, we use an abstract and formal specification as a starting point in our design flow. Based on the synchronous model of computation, we propose a top-down communication refinement approach. This approach decouples the tight global synchronization into process local synchronization, and utilizes synchronizers to achieve process synchronization consistency during refinement. Meanwhile, protocol refinement can be incorporated to satisfy design constraints such as reliability and throughput.

The thesis summarizes the major research results on the three topics.

sted, utgiver, år, opplag, sider
Stockholm: KTH , 2007. , s. xvi, 109
Serie
Trita-ICT-ECS AVH, ISSN 1653-6363 ; 2007:02
Emneord [en]
On-Chip Communication, Network-on-Chip, System-on-Chip
HSV kategori
Identifikatorer
URN: urn:nbn:se:kth:diva-4290ISBN: 978-91-7178-580-0 (tryckt)OAI: oai:DiVA.org:kth-4290DiVA, id: diva2:11666
Disputas
2007-03-15, D, KTH Forum, Isafördsgatan 39, Kista, 13:00 (engelsk)
Opponent
Veileder
Merknad
QC 20100525Tilgjengelig fra: 2007-02-28 Laget: 2007-02-28 Sist oppdatert: 2010-08-06bibliografisk kontrollert
Delarbeid
1. Flit admission in on-chip wormhole-switched networks with virtual channels
Åpne denne publikasjonen i ny fane eller vindu >>Flit admission in on-chip wormhole-switched networks with virtual channels
2004 (engelsk)Inngår i: 2004 INTERNATIONAL SYMPOSIUM ON SYSTEM-ON-CHIP, PROCEEDINGS, IEEE conference proceedings, 2004, s. 21-24Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Flit-admission solutions for wormhole switches must minimize the complexity of the switches in order to achieve cheap implementations. We propose to couple flit-admission buffers with physical channels so that flits from a flit-admission buffer are dedicated to a physical channel. By the coupling strategy, for input-queuing wormhole lane switches, the complexity of the crossbars can be simplified from 2p x p to (p + 1) x p, where p is the number of physical channels; for output-queuing wormhole lane switches, the additional complexity is also minimal. We evaluate the flit-admission solutions derived from the coupling with uniformly distributed random traffic in a 2D mesh network. Experimental results show that these solutions exhibit good performance in terms of latency and throughput.

sted, utgiver, år, opplag, sider
IEEE conference proceedings, 2004
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-13036 (URN)000227185900005 ()2-s2.0-21244466371 (Scopus ID)
Konferanse
International Symposium on System-on-Chip, Tampere, Finland, November 2004.
Merknad

QC 20100524. QC 20160212

Tilgjengelig fra: 2010-05-24 Laget: 2010-05-24 Sist oppdatert: 2016-02-12bibliografisk kontrollert
2. Flit ejection in on-chip wormhole-switched networks with virtual channels
Åpne denne publikasjonen i ny fane eller vindu >>Flit ejection in on-chip wormhole-switched networks with virtual channels
2004 (engelsk)Inngår i: 22ND NORCHIP CONFERENCE, PROCEEDINGS, IEEE conference proceedings, 2004, s. 273-276Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

An ideal it-ejection model is typically assumed in the literature for wormhole switches with virtual channels. With such a model, its are ejected from the network immediately upon reaching their destinations. This achieves optimal performance but is very costly. The required number of sink queues of a switch for absorbing its is p center dot v, where p is the number of physical channels (PCs) of the switch; v the number of lanes per PC To achieve cheap silicon implementations, it-ejection solutions must be cost-effective. We present a novel it-ejection model and a variant of it where the required number of sink queues of a switch is p, i.e., independent of v. We evaluate the it-ejection models with uniformly distributed random traf c in a 2D mesh network. Experimental results show that they exhibit good performance in latency and throughput.

sted, utgiver, år, opplag, sider
IEEE conference proceedings, 2004
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-13037 (URN)10.1109/NORCHP.2004.1423876 (DOI)000227801500068 ()2-s2.0-21244435987 (Scopus ID)
Konferanse
IEEE NorChip Conference, Oslo, Norway, November 2004.
Merknad

QC 20100524. QC 20160212

Tilgjengelig fra: 2010-05-24 Laget: 2010-05-24 Sist oppdatert: 2016-02-12bibliografisk kontrollert
3. Connection-oriented multicasting in wormhole-switched networks on chip
Åpne denne publikasjonen i ny fane eller vindu >>Connection-oriented multicasting in wormhole-switched networks on chip
2006 (engelsk)Inngår i: IEEE Computer Society Annual Symposium on VLSI, Proceedings - EMERGING VLSI TECHNOLOGIES AND ARCHITECTURES, 2006, s. 205-210Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Network-on-Chip (NoC) proposes networks to replace buses as a scalable global communication interconnect for future SoC designs. However, a bus is very efficient in broadcasting. As the system size scales up to explore the chip capacity, broadcasting in NoCs must be efficiently supported. This paper presents a novel multicast scheme in wormhole-switched NoCs. By this scheme, a multicast procedure consists of establishment, communication and release phase. A multicast group can request to reserve virtual channels during establishment and has priority on arbitration of link bandwidth. This multicasting method has been effectively implemented in a mesh network with dead-lock freedom. Our experiments show that the multicast technique improves throughput, and does not exhibit significant impact on unicast performance in a network with mixed unicast and multicast traffic if the network is not saturated.

Emneord
Communication channels (information theory); Interconnection networks; Microprocessor chips; Switching networks; Telecommunication traffic; Throughput; Chip capacity; Global communication interconnect; Multicast traffic; Virtual channels; Multicasting
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-13057 (URN)10.1109/ISVLSI.2006.31 (DOI)000237225200032 ()2-s2.0-33749348991 (Scopus ID)
Konferanse
ISVLSI'2006
Merknad

QC 20100525. QC 20160212

Tilgjengelig fra: 2010-05-25 Laget: 2010-05-25 Sist oppdatert: 2016-02-12bibliografisk kontrollert
4. TDM virtual-circuit configuration for network-on-chip
Åpne denne publikasjonen i ny fane eller vindu >>TDM virtual-circuit configuration for network-on-chip
2008 (engelsk)Inngår i: IEEE Transactions on Very Large Scale Integration (vlsi) Systems, ISSN 1063-8210, E-ISSN 1557-9999, Vol. 16, nr 8, s. 1021-1034Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

In network-on-chip (NoC), time-division-multiplexing (TDM) virtual circuits (VCs) have been proposed to satisfy the quality-of-service requirements of applications. TDM VC is a connection-oriented communication service by which two or more connections take turns to share buffers and link bandwidth using dedicated time slots. In the paper, we first give a formulation of the multinode VC configuration problem for arbitrary NoC topologies. A multinode VC allows multiple source and destination nodes on it. Then we address the two problems of path selection and slot allocation for TDM VC configuration. For the path selection, we use a backtracking algorithm to explore the path diversity, constructively searching the solution space. In the slot allocation phase, overlapped VCs must be configured such that no conflict occurs and their bandwidth requirements are satisfied. We define the concept of a logical network (LN) as an infinite set of associated (time slot, buffer) pairs with respect to a buffer on a given VC. Based on this concept, we develop and prove theorems that constitute sufficient and necessary conditions to establish conflict-free VCs. They are applicable for networks where all nodes operate with the same clock frequency but allowing different phases. Using these theorems, slot allocation for VCs is a procedure of assigning VCs to different LNs. TDM VC configuration can thus be predictable and correct-by-construction. Our experiments on synthetic and real applications validate the effectiveness and efficiency of our approach.

Emneord
Logical network, network-on-chip (NoC), quality of service (QoS), time division multiplexing (TDM), virtual circuit
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-13058 (URN)10.1109/TVLSI.2008.2000673 (DOI)000257987400008 ()2-s2.0-48349101315 (Scopus ID)
Merknad
QC 20100525Tilgjengelig fra: 2010-05-25 Laget: 2010-05-25 Sist oppdatert: 2017-12-12bibliografisk kontrollert
5. Traffic configuration for evaluating networks on chips
Åpne denne publikasjonen i ny fane eller vindu >>Traffic configuration for evaluating networks on chips
2005 (engelsk)Inngår i: Fifth International Workshop on System-on-Chip for Real-Time Applications, Proceedings, IEEE Computer Society, 2005, s. 535-540Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Network-on-Chip (NoC) provides a network as a global communication platform for future SoC designs. Evaluating network architectures requires both synthetic workloads and application-oriented traffic. We present our traffic configuration methods that can be used to configure uniform and locality traffic as synthetic workloads, and to configure channel-based traffic for specific application(s). We also illustrate the significance of applying these methods to configure traffic for network evaluation and system simulation. These traffic configuration methods have been integrated into our Nostrum NoC simulation environment.

sted, utgiver, år, opplag, sider
IEEE Computer Society, 2005
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-13038 (URN)10.1109/IWSOC.2005.107 (DOI)000231591800102 ()2-s2.0-33748913346 (Scopus ID)0-7695-2403-6 (ISBN)
Konferanse
5th International Workshop on System-on- Chip for Real-time Applications, Alberta, Canada, July 2005
Merknad

QC 20100524. QC 20160209

Tilgjengelig fra: 2010-05-24 Laget: 2010-05-24 Sist oppdatert: 2020-03-05bibliografisk kontrollert
6. Connection-oriented multicasting in wormhole-switched networks on chip
Åpne denne publikasjonen i ny fane eller vindu >>Connection-oriented multicasting in wormhole-switched networks on chip
2006 (engelsk)Inngår i: Proceedings of the 16th ACM Great Lakes symposium on VLSI, Association for Computing Machinery (ACM), 2006, s. 296-301Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Deflection routing is being proposed for networks on chips since it is simple and adaptive. A deflection switch can be much smaller and faster than a wormhole or virtual cut-through switch. A deflection-routed network has three orthogonal characteristics: topology, routing algorithm and deflection policy. In this paper we evaluate deflection networks with different topologies such as mesh, torus and Manhattan Street Network, different routing algorithms such as random, dimension XY, delta XY and minimum deflection, as well as different deflection policies such as non-priority, weighted priority and straight-through policies. Our results suggest that the performance of a deflection network is more sensitive to its topology than the other two parameters. It is less sensitive to its routing algorithm, but a routing algorithm should be minimal. A priority-based deflection policy that uses global and history-related criterion can achieve both better average-case and worst-case performance than a non-priority or priority policy that uses local and stateless criterion. These findings are important since they can guide designers to make right decisions on the deflection network architecture, for instance, selecting a routing algorithm or deflection policy which has potentially low cost and high speed for hardware implementation.

sted, utgiver, år, opplag, sider
Association for Computing Machinery (ACM), 2006
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-6841 (URN)2-s2.0-33750928209 (Scopus ID)
Konferanse
Great Lakes Symposium on VLSI
Merknad

QC 20100525. QC 20160212.

Tilgjengelig fra: 2007-02-28 Laget: 2007-02-28 Sist oppdatert: 2016-02-12bibliografisk kontrollert

Open Access i DiVA

fulltekst(494 kB)7758 nedlastinger
Filinformasjon
Fil FULLTEXT01.pdfFilstørrelse 494 kBChecksum MD5
e595b3204d0817a0c1e911066d2c204e1521c379c2475b54a41a9fb2640807b3ffecae36
Type fulltextMimetype application/pdf

Personposter BETA

Lu, Zhonghai

Søk i DiVA

Av forfatter/redaktør
Lu, Zhonghai
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar
Totalt: 7758 nedlastinger
Antall nedlastinger er summen av alle nedlastinger av alle fulltekster. Det kan for eksempel være tidligere versjoner som er ikke lenger tilgjengelige

isbn
urn-nbn

Altmetric

isbn
urn-nbn
Totalt: 2010 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf