--- title: "MeerTRAP GPU cluster" ocid: "ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5" canonical_url: "https://d3tenders.com/contract/?ocid=ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5" markdown_url: "https://d3tenders.com/contract/ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5.md" json_url: "https://d3tenders.com/contract/ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5.json" source: "Contracts Finder" current_stage: "Tender" buyer: "THE UNIVERSITY OF MANCHESTER" published: "2017-12-08" --- # MeerTRAP GPU cluster Buyer: THE UNIVERSITY OF MANCHESTER Current stage: Tender OCID: ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5 [View canonical contract page](https://d3tenders.com/contract/?ocid=ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5) [Download OCDS JSON](https://d3tenders.com/contract/ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5.json) ## Summary The procurement process pertains to the contract titled "MeerTRAP GPU cluster," initiated by The University of Manchester, which is seeking to procure high-performance computing equipment for the MeerTRAP project. This initiative falls within the information systems and servers industry category, located in Manchester, United Kingdom. The procurement method is an open procedure, and the tender is currently in the Tender stage, with the deadline for submission set for 12 January 2018. The contract period is scheduled to commence on 30 January 2018 and conclude on 28 March 2018. This tender presents an exciting opportunity for businesses involved in the provision of advanced computing technologies, particularly those specialising in high-performance servers and data processing systems. Companies that develop GPU clusters or have expertise in real-time data processing solutions may find this contract particularly relevant. Engaging with this tender could enhance growth prospects, as the winning supplier will play a pivotal role in supporting innovative research within the realm of astronomy and data science facilitated by the MeerKAT telescope. ## Notice The MeerTRAP project is supported by the European Research Council, Advanced Grant, awarded to Prof. B.W. Stappers (University of Manchester). The MeeTRAP group consists of four postdoctoral researchers and three PhD students, all placed at the University of Manchester. MeerTRAP will be a real-time, fully-commensal survey for repeating and one-off radio signals originating from pulsars and other fast radio transients using the 64-dish MeerKAT telescope in South Africa. MeerTRAP will piggy-back all the other large surveys performed by MeerKAT, and reuse their data in a completely different way to achieve its goals. Having access to thousands of hours of observing time at its disposal, MeerTRAP must be supported by a high-availability, high-performance GPU cluster to allow for the real-time searching of the incoming data, using optimized GPU pipelines developed by the group. MeerKAT is a precursor telescope for the Square Kilometre Array. The MeerTRAP GPU cluster will be a real-time, commensal processing system hosted in the KAPB, that will process data acquired with the MeerKAT radio telescope in South Africa. MeerKAT consists of 64 dishes and has the ability to look at many different locations ("beams") in the sky simultaneously. When MeerKAT will be observing, a beamformer (the hardware system responsible for creating these beams), will create 400 beams on the sky simultaneously, with each of them being in essence a data stream. The MeerTRAP cluster will receive these 400 data streams and every GPU in it will be responsible for the real time processing of 3 beams (actually, 396 beams on the 132 GPUs expected to populate the cluster)*. The MeerTRAP cluster will consist a leaf node connected to the spine of the MeerKAT multicast network. Additional information: To express an interest in this project please visit the website below where you will need to register to obtain tender documentation. https://in-tendhost.co.uk/universityofmanchester/aspx/Home Funding Type (WEFO or EU) : EU Funding Description The MeerTRAP project is supported by the European Research Council, Advanced Grant, awarded to Prof. B.W. Stappers (University of Manchester). The MeeTRAP group consists of four postdoctoral researchers and three PhD students, all placed at the University of Manchester. MeerTRAP will be a real-time, fully-commensal survey for repeating and one-off radio signals originating from pulsars and other fast radio transients using the 64-dish MeerKAT telescope in South Africa. MeerTRAP will piggy-back all the other large surveys performed by MeerKAT, and reuse their data in a completely different way to achieve its goals. Having access to thousands of hours of observing time at its disposal, MeerTRAP must be supported by a high-availability, high-performance GPU cluster to allow for the real-time searching of the incoming data, using optimized GPU pipelines developed by the group. MeerKAT is a precursor telescope for the Square Kilometre Array. The MeerTRAP GPU cluster will be a real-time, commensal processing system hosted in the KAPB, that will process data acquired with the MeerKAT radio telescope in South Africa. MeerKAT consists of 64 dishes and has the ability to look at many different locations ("beams") in the sky simultaneously. When MeerKAT will be observing, a beamformer (the hardware system responsible for creating these beams), will create 400 beams on the sky simultaneously, with each of them being in essence a data stream. The MeerTRAP cluster will receive these 400 data streams and every GPU in it will be responsible for the real time processing of 3 beams (actually, 396 beams on the 132 GPUs expected to populate the cluster)*. The MeerTRAP cluster will consist a leaf node connected to the spine of the MeerKAT multicast network. Is a Recurrent Procurement Type? : No ## Key Details | Field | Value | | --- | --- | | Publication source | Contracts Finder | | Latest notice | https://www.contractsfinder.service.gov.uk/Notice/d22442e5-61c4-44f6-8a71-18c5c39fcea4 | | Notice type | Tender Notice | | Procurement type | Standard | | Procurement category | Goods | | Procurement method | Open | | Procurement method details | Open procedure (above threshold) | | Tender suitability | Not specified | | Awardee scale | Not specified | | All stages | Tender | ## Dates | Field | Value | | --- | --- | | Publication date | 8 Dec 2017 | | Submission deadline | 12 Jan 2018 | | Future notice date | Not specified | | Award date | Not specified | | Contract period | 30 Jan 2018 - 28 Mar 2018 | | Recurrence | Not specified | ## Values | Field | Value | | --- | --- | | Tender value | Not specified | | Lots value | Not specified | | Awards value | Not specified | | Contracts value | Not specified | ## Status | Field | Value | | --- | --- | | Tender status | Active | | Lots status | Not specified | | Awards status | Not specified | | Contracts status | Not specified | ## Buyer | Field | Value | | --- | --- | | Main buyer | THE UNIVERSITY OF MANCHESTER | | Locality | MANCHESTER | | Post town | Manchester | | Postcode | M13 9PL | | Country | England | | ITL 1 | TLD North West (England) | | ITL 2 | TLD3 Greater Manchester | | ITL 3 | TLD33 Manchester | | Local authority | Manchester | | Electoral ward | Hulme | | Westminster constituency | Manchester Rusholme | | Delivery location | Not specified | ## CPV Codes ### Divisions - 48 - Software package and information systems ### Codes - 48800000 - Information systems and servers ## Release History - 8 Dec 2017 at 13:01 - Tender - Tender Notice - https://www.contractsfinder.service.gov.uk/Notice/d22442e5-61c4-44f6-8a71-18c5c39fcea4 ## Documents - https://www.contractsfinder.service.gov.uk/Notice/d22442e5-61c4-44f6-8a71-18c5c39fcea4 8th December 2017 - Opportunity notice on Contracts Finder - https://in-tendhost.co.uk/universityofmanchester/aspx/ Additional information on how to apply for this contract ## Notice URLs - https://in-tendhost.co.uk/universityofmanchester/aspx/Home ## Provenance This Markdown file is an alternate public rendering of the D3 Tenders contract record. The canonical page is https://d3tenders.com/contract/?ocid=ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5. The underlying structured data is available as OCDS JSON at https://d3tenders.com/contract/ocds-b5fd17-05fe9bff-0827-4de1-9365-88197f5614a5.json.