Status

• 29.10.2008 - notification to sites about the results sent
• 02.10.2008 - evaluation accepted by AMB, sent to PMB for final approval
• 24.09.2008 - Evaluation committee meeting
• 19.09.2008 - Deadline for offers
• 09.09.2008 - Call for offers sent to Collaboration Board
• Official document: https://edms.cern.ch/file/936492/1.6/EGEE-III-SA1-TEC-936492-SeedResources-v1_6.pdf

Agreement with sites according to amount of provided resources

• GRIF: 39 cores, 4.5 TB
• STFC: 146 cores, 1 TB
• CYFRONET: 30 cores, 15 TB
• GR-01-AUTH: 42 cores, 6.5 TB

Summary: 257 cores and 27 TB in total

Requirements for tender

draft version 0.1

General:

1. Money should be allocated to two or three sites which have experience in operation (including grid core services for VO) and agree to follow the requirements defined below.
2. Declared resources provided by those sites constitute so called central-pool of seed resources.
3. Winning offers should guarantee:
   • substantial number of CPUs, storage and a full set of core services,
   • possibility to set-up the pool within one month after announcing results of this tender (it is crucial to be independent from won money transfer and hardware acquire procedures).
4. Money won in the tender should be spent on hardware only (not personnel), but sites need to provide relevant man-power for maintaining services and collaborate on operation of seed-resources pool.
5. The resources will be used to be made available to new user communities not linked to the partners of the EGEE consortium.

Requirements for sites:

1. The site must be a full EGEE production site, already running both site services and core services.
2. The site must be able to delegate a representative to Resource Allocation Group.
3. The site must agree to act according procedures that will be worked out by Resource Allocation Group concerning communication with VOs, configuration, reporting usage, providing feedback, etc.
4. The site must specify resources that would be made available as central-pool of seed-resources:
   • number of CPUs/cores (number of slots per kEUR won in tender),
   • power of single CPUs/core (in kSpecInt per kEUR won in tender),
   • capacity of offered storage (in TB per KEUR won in tender):
5. The site must be ready to provide on request each of the following services:
   • UI
   • WMS
   • LFC
   • FTS
   • VOMS
   • MyProxy
   • AMGA (if included in official release)
   • HYDRA (if included in official release)
6. The site must be ready to set-up all activities related to central-pool of seed resources withing one month after publication of the tender results.
7. The resources will be made available to VOs on short notice after they are requested (e.g. max 3 days for site services, max 1 week for core services).
8. The sites must support the MPI and Short Deadline Job configurations that are important for certain classes of applications.
9. Following SLD metrics should be kept:
   • minimum site availability: 80%
   • Minimum site reliability: 90%
   • Maximum time to acknowledge GGUS tickets: four hours within site administration working time
   • Maximum time to resolve GGUS incidents: five working days

Templates for tender offer (draft)

Contact data:

1. Name of the site:
2. Responsible person:
3. Contact e-mail:

1. Start-up time after the results of tender will be publish, how much time the site need to set-up activity (in days, specify conditions if any):
2. Power of offered CPUs (in kSPECInt per kEUR win in tender):
3. Capacity of offered storage (in TB per KEUR win in tender):
   • Note that, resources offered are understood as pool of resources that Resource Allocation Group can allocate to VOs according their needs.
4. Ability to maintain services for VOs using seed resources:
   • UI (yes/no):
   • WMS (yes/no):
   • LFC (yes/no):
   • FTS (yes/no):
   • VOMS (yes/no):
   • MyProxy (yes/no):
   • AMGA (yes/no):
   • HYDRA (yes/no):
5. Time needed for VO reconfiguration, i.e. enabling new VOs (in days):
6. Time needed for ticket response (in hours):
7. Time needed for solving ticket (in days):
   • Note that we refer hours within working hours as declare in GOCDB and days in working days.
8. Minimal amount of money win in tender from which the offer is valid (in KEUR):

Obligatory requirements:

1. The full EGEE production site must be maintained, including CE, SE and set of working nodes.
2. The participating sites must support the MPI and Short Deadline Job configurations that are important for certain classes of applications
3. The site must be able to send representative of the site to Resource Allocation Group
4. The site must agree to act according procedures that will be work out by Resource Allocation Group concerning communication with VOs, configuration, reporting usage, providing feedback, etc.

Intro to Central Pool of Seed Resources

Goals The na4.vo.eu-egee.org virtual organization brings together resources-- computing and human--to help the NA4 activity provide better support to the EGEE user community. Specificially, people within this VO will:

• help new users port their applications to the EGEE production infrastructure,
• validate proposed solutions to problems facing users,
• validate documentation/FAQs, and
• write SAM tests to verify certain critical functionalities.

Those who are porting their software to the grid are expected to migrate to another VO after the porting and initial validation phases; these phases should not last more than 6 months. The resources within the VO are expected to cover the full range of grid functionality and use of them will help new users better understand what services must be supported within their own, new VOs.

Users

The goals require that two groups of people must have access to resources in this VO: 1) end users that are porting applications with the help of the GASuC team and 2) support staff from the NA4 activity. The end users' memberships will be for a maximum of 6 months. NA4 support staff may remain in the VO indefinitely.

Requirements for Sites

Because porting of new or innovative applications is a large part of the work that will be carried out within this VO, it is essential that the participating sites provide the full range of services in gLite and those additional services/configurations deemed important by NA4 management.

In particular the full functional range of the EGEE production infrastructure must be covered: CE, SE, WMS, LFC, FTS, MyProxy, VOMS, and a UI. Also support should be provided for services like AMGA and HYDRA that are close to certification. Equally important, the participating sites must support the MPI and Short Deadline Job configurations that are important for certain classes of applications.

AUP

This Acceptable Use Policy applies to all members of the na4.vo.eu-egee.org Virtual Organisation, hereafter referred to as the VO, with reference to use of the LCG/EGEE Grid infrastructure, hereafter referred to as the Grid. The Centre National de la Recherche Scientifique (CNRS) owns and gives authority to this policy. The goal of the VO is facilitate the porting of applications to the grid infrastructure and improve the support provided by the NA4 activity of EGEE. Members and Managers of the VO agree to be bound by the Grid Acceptable Usage Rules, VO Security Policy and other relevant Grid Policies, and to use the Grid only in the furtherance of the stated goal of the VO.