PapersFlow Research Brief

Physical Sciences · Computer Science

Interconnection Networks and Systems
Research Guide

What is Interconnection Networks and Systems?

Interconnection Networks and Systems is the field focused on the design, architecture, and optimization of Networks on Chip (NoC) within System-on-Chip (SoC) designs, encompassing routing algorithms, performance evaluation, power optimization, wireless interconnects, fault tolerance, and multi-core processors.

This field addresses interconnection networks central to multi-core processors and SoC architectures, with 69,895 works published. Key areas include NoC architecture, routing algorithms, and power optimization. Benchmarks like SPLASH-2 and PARSEC evaluate performance in shared-memory multiprocessors.

Topic Hierarchy

100%
graph TD D["Physical Sciences"] F["Computer Science"] S["Computer Networks and Communications"] T["Interconnection Networks and Systems"] D --> F F --> S S --> T style T fill:#DC5238,stroke:#c4452e,stroke-width:2px
Scroll to zoom • Drag to pan
69.9K
Papers
N/A
5yr Growth
724.7K
Total Citations

Research Sub-Topics

Why It Matters

Interconnection networks enable efficient communication in multi-core processors and SoCs used in telecommunications, multimedia, and consumer electronics. "Networks on chips: a new SoC paradigm" by Benini and De Micheli (2002) describes NoCs as solutions to time-to-market pressures in these domains, supporting complex electronic engines. Benchmarks such as "The SPLASH-2 programs" by Woo et al. (1995) and "The PARSEC benchmark suite" by Bienia et al. (2008) quantify properties like data sharing and locality in parallel applications, aiding optimization for real systems. "Principles and Practices of Interconnection Networks" by Dally and Towles (2004) addresses communication challenges in digital systems with economical network solutions pervasive across components.

Reading Guide

Where to Start

"Networks on chips: a new SoC paradigm" by Benini and De Micheli (2002), as it introduces NoC as a core paradigm for SoC communication challenges in practical domains.

Key Papers Explained

"Networks on chips: a new SoC paradigm" by Benini and De Micheli (2002) establishes NoC foundations, which "Principles and Practices of Interconnection Networks" by Dally and Towles (2004) expands with design principles for digital systems. Benchmarks in "The SPLASH-2 programs" by Woo et al. (1995) and "The PARSEC benchmark suite" by Bienia et al. (2008) build evaluation frameworks on these architectures. Graph methods from "A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs" by Karypis and Kumar (1998) support topology optimization across them.

Paper Timeline

100%
graph LR P0["Shortest Connection Networks And...
1957 · 4.5K cites"] P1["Validity of the single processor...
1967 · 4.2K cites"] P2["A bridging model for parallel co...
1990 · 3.7K cites"] P3["A generalized processor sharing ...
1993 · 3.7K cites"] P4["A Fast and High Quality Multilev...
1998 · 5.6K cites"] P5["On power-law relationships of th...
1999 · 4.2K cites"] P6["Networks on chips: a new SoC par...
2002 · 3.7K cites"] P0 --> P1 P1 --> P2 P2 --> P3 P3 --> P4 P4 --> P5 P5 --> P6 style P4 fill:#DC5238,stroke:#c4452e,stroke-width:2px
Scroll to zoom • Drag to pan

Most-cited paper highlighted in red. Papers ordered chronologically.

Advanced Directions

Current work builds on NoC power optimization and fault tolerance, extending Dally and Towles (2004) principles to multi-core processors amid 69,895 papers. Benchmarks like PARSEC (Benia et al., 2008) guide CMP studies, with open questions in dynamic routing and wireless interconnects.

Papers at a Glance

# Paper Year Venue Citations Open Access
1 A Fast and High Quality Multilevel Scheme for Partitioning Irr... 1998 SIAM Journal on Scient... 5.6K
2 Shortest Connection Networks And Some Generalizations 1957 Bell System Technical ... 4.5K
3 On power-law relationships of the Internet topology 1999 4.2K
4 Validity of the single processor approach to achieving large s... 1967 4.2K
5 A generalized processor sharing approach to flow control in in... 1993 IEEE/ACM Transactions ... 3.7K
6 Networks on chips: a new SoC paradigm 2002 Computer 3.7K
7 A bridging model for parallel computation 1990 Communications of the ACM 3.7K
8 The SPLASH-2 programs 1995 3.6K
9 The PARSEC benchmark suite 2008 3.4K
10 Principles and Practices of Interconnection Networks 2004 3.3K

Frequently Asked Questions

What are Networks on Chip in interconnection networks?

Networks on Chip (NoC) provide an integrated communication solution for System-on-Chip (SoC) designs in telecommunications, multimedia, and consumer electronics. "Networks on chips: a new SoC paradigm" by Benini and De Micheli (2002) explains that NoCs address design challenges under time-to-market constraints. They enable complex electronic engines through optimized on-chip interconnects.

How do graph partitioning algorithms apply to interconnection networks?

Graph partitioning reduces graph size by collapsing vertices and edges, partitions the smaller graph, and uncoarsens to optimize irregular graphs. "A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs" by Karypis and Kumar (1998) presents a multilevel scheme for this process. It supports efficient routing and topology design in NoC architectures.

What role do benchmarks play in evaluating interconnection networks?

"The SPLASH-2 programs" by Woo et al. (1995) characterize parallel applications for shared-address-space multiprocessors, measuring data sharing and locality. "The PARSEC benchmark suite" by Bienia et al. (2008) targets Chip-Multiprocessors (CMPs) with diverse workloads beyond high-performance computing. These suites facilitate performance evaluation of NoC designs.

What are key principles of interconnection networks?

Interconnection networks optimize communication between digital system components economically. "Principles and Practices of Interconnection Networks" by Dally and Towles (2004) covers design challenges and solutions pervasive in modern systems. They address the communication crisis in multi-core and SoC environments.

How does generalized processor sharing apply to network flow control?

Generalized processor sharing (GPS) allocates resources in rate-based flow control for integrated services networks. "A generalized processor sharing approach to flow control in integrated services networks: the single-node case" by Parekh and Gallager (1993) shows GPS provides fair service in virtual circuit packet networks. It models single-node behavior for broader network optimization.

What is the significance of shortest connection networks?

Shortest connection networks interconnect terminals with minimal direct links. "Shortest Connection Networks And Some Generalizations" by Prim (1957) provides graphical and computational procedures for solutions. These methods extend to multi-terminal generalizations relevant to NoC topologies.

Open Research Questions

  • ? How can multilevel graph partitioning be adapted for dynamic NoC topologies in multi-core processors?
  • ? What power optimization techniques best balance latency and energy in wireless interconnects for SoCs?
  • ? How do fault-tolerant routing algorithms maintain performance under varying failure rates in large-scale NoCs?
  • ? Which bridging models extend Valiant's parallel computation framework to modern CMP interconnection networks?
  • ? How do power-law topologies from Internet studies inform scalable NoC architectures for exascale systems?

Research Interconnection Networks and Systems with AI

PapersFlow provides specialized AI tools for Computer Science researchers. Here are the most relevant for this topic:

See how researchers in Computer Science & AI use PapersFlow

Field-specific workflows, example queries, and use cases.

Computer Science & AI Guide

Start Researching Interconnection Networks and Systems with AI

Search 474M+ papers, run AI-powered literature reviews, and write with integrated citations — all in one workspace.

See how PapersFlow works for Computer Science researchers