DATA MODIFICATION(S) Sample Clauses

DATA MODIFICATION(S). Any modification to the Purpose, Justification, Description of Data to be Shared/Data Licensing Statement(s), and Permissible Use, is required to be approved through HCA’s Data Request Process. Contractor must notify HCA’s Contract Manager of any requested changes to the Data elements, Use, records linking needs, research needs, and any other changes from this Contract, immediately to start the review process. Approved changes will be documented in an Amendment to the Contract.

DATA MODIFICATION(S). The data structures used by a program depend largely on the algorithm used, but within a given algorithm there may be changes that have to be made to fit a particular programming model or in order to achieve acceptable performance. Scalar types The underlying data type may need to be altered to improve efficiency on a parallel processor. This is especially true if double-precision floating-point values are used because a number of massively-parallel architectures do not support them or support them with limited performance. The algorithm may have to be modified to cope with the reduced accuracy of single-precision floating-point. Conversely, some algorithms may be able to accept the reduced accuracy provided by half-precision floating-point or fixed point numbers which may provide better performance on a parallel processor. It may also be necessary to pay more attention to the use of unsigned data types as these may perform differently from signed data types. Structures of Arrays Many algorithms work on structured data, where a number of values are grouped into a structure which is then replicated in a large array. This arrangement is known as an Array of Structures (AOS) and is efficient on a scalar processor where the single thread is operating on each array element in turn. AOS is not so efficient where multiple threads are operating in parallel on different elements of the array. In this case the data is more efficiently stored by having separate arrays for each element of the structure. This arrangement is known as a Structure of Arrays (SOA) and is particularly effective on SIMD architectures where multiple values can be read and written from sequential memory. With very large data sets the separation between related data values in AOS can cause problems with caching and TLBs. In this case it can be effective to limit the size of an SOA to fit within the page size and then replicate this multiple times, leading to an Array of Structures of Arrays (AOSOA). In each case the most efficient partitioning between structures and arrays will depend on the particular architecture being targeted. Image formats Algorithms that process pixel data can choose between the usual interleaved formats or planar formats where the data for each channel is in a separate area of memory. The trade-offs are the same as for the AOS vs. SOA choice for non-pixel data. The 2D nature of image data makes other data arrangements even more efficient for parallel processing. In parti...