LCG 3D Requirement Specification Spreadsheets

As part of the LCG 3D project all participating experiments and software project are invited to specify their functional and storage resource requirements for data which is stored in a relational database and will require a LCG provided service during their production activities in 2005 .

This information collected will be used to:

Description of Individual Spreadsheet Columns

Application/Data Type

The propose spreadsheet contains for each "Application" or "Data Type" a set of rows which summarize the data distribution through the different LCG tiers including the data volume, access and distribution latency expectations of the experiment or software provider.

Each Application or Data Type in the spreadsheet corresponds to a consistent set of tables which are deployed together and are distributed together without modification. In case data is aggregated or transformed in a different way during the distribution from one tier to another, a new data type or application entry should be made. E.g. "raw slow control" data may be generated online and flow up to tier 0, but only "slow control summaries" (a aggregated subset of the same data) may flow further down-stream to tier 1 and tier 2.



Main distribution pattern or "none" in case that particular data is hosted at one tier but not distributed. Please enter also data which is not distributed in order to allow for proper volume and client access estimates at each tier.


Starts a description of the requirements for a particular tier. The value should be "online" for data in the online database or the tier number (0/1/2/3/4) for grid tiers.


Either the source tier from which the data is obtained or the program which creates the new data at this tier.


The total amount of data at this tier which will be stored in 2005.

Number of clients

The maximum number of connecting clients at this tier to this data type.

Access Mode

The list of required access modes at this tier. Eg "r" for data which is only read, "wu" for data for which new table rows are created at this tier and old table rows are updated.


The ownership model for this data. Is the data owned (created and updated) by

Write/Update Rate

The maximum rate (in MB/day) of new or changed table row data.

Maximum Distribution Latency

The maximum time since creation or update at the source tier after which the data need to be available at the destination tier.

RAL used

Do all applications which maintain the data use the Relational Abstraction Layer?

Oracle Implementation

Does an Oracle implementation exist for the applications which maintain this data?

MySQL Implementation

Does an MySQL implementation exist for the applications which maintain this data?


Email of the person who is responsible for the requirements in this section. In case of several different responsible persons for functional and volume requirements please list all.