Breaking the petaflops barrier

Grice, Don; Brandt, H.; Wright, Christopher S.; McCarthy, P.; Emerich, A.; Schimke, T.; Archer, Charles J; Carey, J.; Sanders, P. P.; Fritzjunker, J. A.; Lewis, Scott M.; Germann, P.

doi:10.1147/jrd.2009.5429067

Cited by 34 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the viscous monomer for the cladding is coated on a substrate (1). Next, the viscous monomer for core is dispensed into the cladding layer by inserting the bottom of the needle in the cladding layer, and then the needle is scanned horizontally as shown in the inset (2). Finally both core and cladding are cured under UV exposure followed by post baking at 100 o C (3).…”

Section: The Mosquito Methodsmentioning

confidence: 99%

“…(1) (2) where r is the distance from the core center to the measuring point, n 1 and n 2 are the refractive indexes at the core center (r = 0) and the cladding, respectively, a is the core radius, and g is the index exponent. The g-parameters estimated from the best-fitted curves to the profiles are 2.0 and 4.0 for waveguides A and B, respectively.…”

Section: Refractive-index Distributionmentioning

confidence: 99%

“…In some current HPCs, electrical interconnection is still utilized in board to board data transmissions. However, for saving the power dissipation in those HPC systems, optical interconnection technologies have been expected to be deployed even for board-level wirings [2]. For the board-level interconnects, multimode parallel polymer optical waveguides (PPOWs) have attracted much attention because of their low fabrication cost and high processability [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Graded-index core polymer optical waveguide circuit fabricated using a microdispenser for high-density on-board optical interconnects

Soma

Ishigure

2013

SPIE Proceedings

View full text Add to dashboard Cite

We introduce a simple photo-mask-free fabrication method of multimode polymer optical waveguides (PPOWs): "the Mosquito method" that utilizes a micro-dispenser. In the Mosquito method, a viscous monomer for cores is directly dispensed from a thin needle into a cladding monomer that is coated on a substrate. Because of the monomer diffusion, graded-index (GI) circular core is formed. By adjusting several parameters to control the core shape and diameter, 15-cm long, 40-m core diameter, with 125-m pitch GI-circular core PPOWs (12 channels) are successfully fabricated by the Mosquito method. We experimentally demonstrate superior optical properties of the GI-circular core polymer waveguides: connection loss with GI multimode fibers, inter-channel crosstalk, etc. to SI-square core polymer waveguides composed of the same silicone polymer materials.Furthermore, we fabricate GI-core PPOW circuits in which perpendicularly curved waveguides are involved by applying the Mosquito method. The curved PPOWs with a 250-m pitch are obtained successfully and the curve-radius is varied from 3 mm to 20 mm by adjusting the scanning program of the needle. The insertion loss of the curved waveguides is measured for evaluating the bending losses. The waveguide with lager curve-radius shows lower insertion loss, namely low bending loss. Additionally, it is experimentally confirmed that the curved waveguides with higher numerical aperture (NA = 0.30) fabricated utilizing a different silicone resin for the cladding exhibit a bending loss as low as 0.5 dB even under 3-mm curve radius. The Mosquito-method would be a promising method for realizing highspeed and high-density on-board optical interconnects.

show abstract

Section: The Mosquito Methodsmentioning

confidence: 99%

Section: Refractive-index Distributionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Graded-index core polymer optical waveguide circuit fabricated using a microdispenser for high-density on-board optical interconnects

Soma

Ishigure

2013

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…In 2008, Roadrunner [1], the first petaflop supercomputer (i.e., >10 15 floating point operations per second) was constructed. The large number of microprocessor cores in such petascale machines must be interconnected with a high capacity communications network to permit efficient computation.…”

Section: Optics In Petascale Supercomputersmentioning

confidence: 99%

Optical Interconnects in Future Servers

Kash

Benner²,

Doany

et al. 2011

Optical Fiber Communication Conference/National Fiber Optic Engineers Conference 2011

View full text Add to dashboard Cite

Optical interconnects are common in today's petascale supercomputers, and will become pervasive at the exascale during this decade. Technologies that can meet the challenging technological and economic requirements for the exascale will be reviewed. Optics in Petascale SupercomputersIn 2008, Roadrunner[1], the first petaflop supercomputer (i.e., >10 15 floating point operations per second) was constructed. The large number of microprocessor cores in such petascale machines must be interconnected with a high capacity communications network to permit efficient computation. Higher interconnect bandwidth will generally result in more efficient use of the microprocessors in real calculations. Prior to 2005, this network used electrical interconnects in essentially all supercomputers. Communication bitrates increased beyond ~2Gb/s starting in about 2005. For distances greater than about 20m, electrical interconnects were impractical and optics began to be used for these longer rack-to-rack interconnects (e.g., ASCI Purple[2]). For cost reasons, however, the shorter rack-to-rack interconnect in this machine was still electrical. By 2008, communication bitrates had increased to 5Gb/s (InfiniBand DDR), and the practical reach of electrical interconnects was shorter. At the same time, the cost of optics had decreased. As a result, for Roadrunner, all node-to-node communication was optical. However, in these examples, all the intra-node communication was electrical. Also, the conversion to optics was at the edge of the node/rack, requiring an electrical interconnect from the chip through several levels of packaging before conversion to photons. For machines with peak performance beyond 10PF, optics will play a far more important role. For example, in the planned POWER7-IH systems [3], all the board-to-board interconnects, will be optical. A major change in the packaging tightly integrates the optics onto the chip-level package. Optical transmitter and receiver modules are placed on the same ceramic Hub (switch) Module [4,5] , with each channel using a 10Gb/s signaling rate and 8b10b coding. Projections for future supercomputersBecause microprocessor clock speeds are not rapidly increasing, supercomputing performance improvements come from larger numbers of processing cores operating in parallel. For example [7], the #1 machine on the top500 list from June, 2002 was the NEC Earth-Simulator with 5120 single core processor chips, as compared to Roadrunner which has some 97,920 cores in 12,240 Cell Broadband Engine® processor chips plus 6,562 dual-core AMD Opteron® chips [8]. This increasing core and chip count drives increasing communication bandwidth at higher bitrates.Trends for optical interconnects in future supercomputers can be extrapolated from past growth. Historically, computational power has grown approximately 10x every 4 years [7], and this growth is expected to continue. Affordable supercomputing requires that the cost and power consumption of supercomputers grow more slowly than computational power. ...

show abstract

“…1 For supercomputers to continue historical performance trends: reaching exa-scale (1,000 PFlop/s) performance within the next decade, more power-efficient and higher-BW optical interconnects must be developed. These next-generation parallel optical modules also must offer improved density and lower cost compared to today's offerings.…”

Section: Introductionmentioning

confidence: 99%

Development of high-speed VCSELs beyond 10 Gb/s at Emcore

Xie

Lei

et al. 2010

SPIE Proceedings

View full text Add to dashboard Cite

We report developments at Emcore on serial 850 nm vertical-cavity surface-emitting lasers (VCSELs) operated up to 25 Gb/s. They have been designed to provide a solution not only to meet stringent 10 Gb/s IEEE and Fiber Channel specifications but also for emerging demands of 17 Gb/s Fiber Channel serial and 100 Gb/s (4x25 Gb/s or 5x20 Gb/s) parallel applications in local and storage area networks. This paper covers 10 Gb/s GenX production distributions and improved GenX VCSEL device design to meet low-power requirements at 20 Gb/s. We have successfully demonstrated low threshold current of 0.65 mA at 25°C using nominal 7.3 μm oxide-aperture GenX VCSELs. They can be directly modulated up to 25 Gb/s with open eyes at 6 mA bias. With the same design, open eyes of 20 Gb/s is achieved at bias current as low as 4 mA (9 KA/cm 2 ) at 25°C and 8 mA (18 KA/cm 2 ) at 70°C. These operation conditions are comparable to current 10 Gb/s GenX VCSELs in production which have been shown a great field history.

show abstract

Breaking the petaflops barrier

Cited by 34 publications

References 6 publications

Graded-index core polymer optical waveguide circuit fabricated using a microdispenser for high-density on-board optical interconnects

Graded-index core polymer optical waveguide circuit fabricated using a microdispenser for high-density on-board optical interconnects

Optical Interconnects in Future Servers

Development of high-speed VCSELs beyond 10 Gb/s at Emcore

Contact Info

Product

Resources

About