Scaling internet routers using optics

Keslassy, Isaac; Chuang, Shang-Tse; Yu, Kun-Ming; Miller, David A. B.; Horowitz, Mark; Solgaard, Olav; McKeown, Nick

doi:10.1145/863955.863978

Cited by 191 publications

(87 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…But limitations of PPS are as follows: 1) centralized scheduling can hardly be implemented because of the complexity in communication; and 2) distributed scheduling will incur complicated out-of-order packets and cell reassembly operation at output port. The loadbalanced switch architecture is based on two identical stages of switches [13] . The first stage of switch dispatches input traffic evenly among central switch models and the second stage of switch sends packets to their destination outputs.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

MPFS: A truly scalable router architecture for next generation Internet

Sun

Dai

Zhang

2008

Sci. China Ser. F-Inf. Sci.

View full text Add to dashboard Cite

A new generation architecture of IP routers called massive parallel forwarding and switching (MPFS) is proposed, which is totally different from modern routers. The basic idea of MPFS is mapping complicated forwarding process into multilevel scalable switch fabric so as to implement packet forwarding in a pipelining and distributed way. This processing mechanism is named forwarding in switching (FIS). By interconnecting multi-stage, lower speed components, called forwarding and switching nodes (FSN), MPFS achieves better scalability in forwarding and switching performance just like MPP. We put emphasis upon IPv6 lookup problem in MPFS and propose a method for partitioning IPv6 FIB and mapping them to switch fabric. Simulation and computation results suggest that MPFS routers can support line-speed forwarding with a million of IPv6 prefixes at 40 Gbps. We also propose an implementation of 160 Tbps core router based on MPFS architecture at last.IP routers, forwarding and switching, scalability, architecture

show abstract

Section: Related Workmentioning

confidence: 99%

“…Considering memory space and lookup FIB performance, we use mapping method of Table 2. We adopt initial array, bitmap splitting, skip nodes and CAM optimization for end nodes and internal nodes [13] . Only CAM nodes save the next hop index of prefixes and the next hop index of other nodes are saved in result nodes.…”

Section: Performance Evaluationmentioning

confidence: 99%

MPFS: A truly scalable router architecture for next generation Internet

Sun

Dai

Zhang

2008

Sci. China Ser. F-Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…An optical two-stage parallel packet switch architecture typically uses an optical switch fabric (e.g., AWGR, tunable laser/filter, and so on). It was proposed in the 100T optical router project of Stanford University [11]. The PFR architecture can guarantee 100% throughput for all traffic and it reduces the implementation complexity of switching fabrics, but it results in the problem of packet mis-sequencing and the scaling problem of multi-plane interconnection.…”

Section: Multi-plane Parallel Forwarding Optical Router (Pfr)mentioning

confidence: 99%

High-Performance Hybrid-Switching Optical Router for IP over WDM Integration

et al. 2005

View full text Add to dashboard Cite

Because pure electrical routers with their bandwidth limitations can hardly keep up with the tremendous traffic growth in the Internet, optical routers based on various optical switching techniques including optical wavelength switching (OWS), optical burst switching (OBS), and optical packet switching (OPS) have been suggested to cope with this problem. However, because OBS and OPS are both in their early experimental phase and OWS only provides coarse granularity switching, a hybrid-switching optical router with combined OWS and electrical packet switching is a necessity in order to accommodate the entire multi-granularity traffic with multiservice requirements in a cost-effective manner. Its coordination capability of optical circuit switching and electrical packet switching enables efficient/intelligent usage of network resources. In this paper, we first review research and developments of such IP routers employing optical switching/interconnection techniques and examine how these techniques can be used inside routers to scale node capacity and to improve optical Internet performance. We also present and study the performance of a terabit optical router with an optical-electrical hybrid-switching fabric. The node architecture is based on the idea of IP over WDM integration with Generalized Multi-Protocol Label Switching (GMPLS). The network-level performance evaluations show that the proposed hybrid-switching optical router is a cost-effective solution for building the next generation GMPLS-based multi-granularity optical Internet.

show abstract

“…One of the most important breakthroughs in switch architectures to overcome the problems of conflict resolution is the load balanced Birkhoff-von Neumann switch [9][10][11][12][13][14][15]. It is proposed to resolve the memory access conflict and yield high throughput without the need of communication and computation overheads.…”

Section: Introductionmentioning

confidence: 99%

An 8 × 8 20 Gbps Reconfigurable Load Balanced TDM Switch IC for High-Speed Networking

Chiu¹,

Hsu

Wu³

et al. 2010

J Sign Process Syst

View full text Add to dashboard Cite

In this paper, we propose a reconfigurable load balanced symmetric TDM switch fabric. We fold this twostage switch to reduce 50% hardware complexity, and then implement a 3.65 mm×3.57 mm prototype switch fabric IC, including a digital 8×8 switch core, eight 16B20B CODECs, eight SERDES ports, eight CML I/O interfaces and a PLL, in 0.18 μm CMOS technology. The digital 8× 8 switch core has reconfigurable connection patterns for the ease of scaling up to an N×N switch (N is power of 4). We propose the 16B20B CODEC scheme to reduce the switch core clock rate by half. In the SERDES, we employ the half-rate scheme and then use static CMOS gates for the low power consumption. We develop a low power, areaefficient and wide-band CML I/O interface with our patented PMOS active load inductive-peaking scheme for high-speed data transmission. With the 16B20B CODEC, the half-rate, and the PMOS active load schemes, almost 50% of the power is saved as compared with the design of the 8B10B CODEC, the full-rate and on-chip inductors CML schemes. Our measurement shows that an 8×8 switch fabric IC can achieve 20 Gbps switching rate and consumes only about 690 mW power. A terabit switch fabric can then be constructed by cascading the designed switch ICs.

show abstract

Scaling internet routers using optics

Cited by 191 publications

References 14 publications

MPFS: A truly scalable router architecture for next generation Internet

MPFS: A truly scalable router architecture for next generation Internet

High-Performance Hybrid-Switching Optical Router for IP over WDM Integration

An 8 × 8 20 Gbps Reconfigurable Load Balanced TDM Switch IC for High-Speed Networking

Contact Info

Product

Resources

About