COMPUTE ATLAS Supercomputer supply-chain graph
1581 systems 558 sites Sourced data

Supercomputers/Systems/AVIDD

USA · operational · hpc

AVIDD

RSS for this system Your watchlist

Also known as AVIDD-B+AVIDD-I; Indiana University AVIDD cluster

Operated by Indiana University .

Researched with AI assistance, checked by scripts. These rows were researched by AI agents reading public operator, laboratory and vendor sources, appended by scripts and checked by automated validation; no subject-matter expert has reviewed them (how this site is made). A mechanical second-reader pass has since fetched every cited source: 1350 of 1581 systems have a readable citation that names them, and 64 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.

Measured performance

Rmax
1.06 TFlop/s
HPL measured
Rpeak
1.84 TFlop/s
Rmax ÷ Rpeak
57.4%
Cores
384
Power
-
Per watt
- GF/W

Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.

How it compares

Ranked against the other machines in this dataset that report the same figure: those that first came into service within two years of AVIDD, and those in USA. Computed from our tables; a metric is ranked only when at least eight peers report it.

MetricValueAgainst its eraAgainst its country
Rmax1.06 TFlop/s#51 of 60 · bottom 16%#316 of 332 · bottom 6%
Rmax ÷ Rpeak57.4 %#47 of 60 · bottom 23%#292 of 331 · bottom 13%

Every metric is ranked largest-first, so a high power-draw rank means a big power draw, not a good one. Rmax ÷ Rpeak uses HPL-measured results only. Rankings move as machines are added; they describe this dataset, not the world.

What this machine is made of

One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.

Role Part Supplier at build Quantity Confidence Source
Integrator -
IBM xSeries cluster Xeon 2.4 GHz (AVIDD-B + AVIDD-I)
 IBM
IBMNYSE
- Reported top500.orgscholarworks.iu.edu
CPU -
Intel Pentium 4 Xeon (Prestonia) 2.4 GHz, 384 benchmarked processors
 Intel
INTCNASDAQ
384
reported
 Reported top500.orgscholarworks.iu.edu
Interconnect -
Myrinet
 Myricom
- Reported top500.org

The read

IBM Xeon Linux cluster of the NSF-funded AVIDD facility (Analysis and Visualization of Instrument-Driven Data) at Indiana University, listed by TOP500 as AVIDD-B plus AVIDD-I, the Bloomington and Indianapolis halves: 384 Pentium 4 Xeon 2.4 GHz cores, Myrinet and Force10 networking, Rmax 1,058 and Rpeak 1,843 GFlop/s, rank 50 in June 2003, 114 in November 2003, 218 in June 2004 and 337 in November 2004 (https://www.top500.org/system/172590/). The project's final report to NSF (award EIA-0116050, 1 September 2001 to 31 August 2003) says AVIDD is a quartet of IBM Linux clusters with 470 processors and 2.2 TFlop/s aggregate peak: two clusters of 208 Prestonia 2.4 GHz processors each (the Bloomington and Indianapolis sites), a 36-processor 64-bit McKinley cluster for computer science research, and an 18-processor 1.3 GHz Pentium III cluster at IU Northwest, plus 10 TB of disk. It states that AVIDD ranked 50th on the June 2003 Top500, the highest-ranked distributed Linux cluster on that list and the first distributed Linux cluster to document more than 1 TFlop/s on Linpack. The TOP500 core count (384) is lower than the 416 Xeon processors the report counts for the two main clusters, so only the benchmarked portion is in the flops figures. IU's later machine is big-red.

Timeline

Public list appearances

Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.

Change history

Nearby in scale

The machines closest to AVIDD by Rmax, smaller to larger.

Same operator

Further reading

Operating a supercomputer

Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.

Inside a node

Latency-optimised versus throughput-optimised processors, NUMA, and the end of the host-to-device copy.

The interconnect

Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.