The Experts below are selected from a list of 1087086 Experts worldwide ranked by ideXlab platform

Mannem Manaswi - One of the best experts on this subject based on the ideXlab platform.

  • A Novel GPU Algorithm for Indexing Columnar Databases with Column Imprints
    2020
    Co-Authors: Mannem Manaswi
    Abstract:

    University of Minnesota M.S. thesis. July 2020. Major: Computer Science. Advisor: ELEAZAR LEAL. 1 computer file (PDF); ix, 49 pages.Columnar database management systems (CDBMS) are specialized database systems that store data in Column-Major Order, i.e. where all the values of each attribute are stored consecutively in storage, as opposed to row-major Order, which stores all the values of each row consecutively and is the most commonly strategy used by relational database systems such as Oracle, SQLServer, PostgreSQL, etc. The Column-Major Order approach makes CDBMS more appropriate for data warehouses because query workloads for the latter systems usually involve retrieving all the values of a small subset of attributes. \par Just like in relational database management systems, query response time performance in CDBMS can greatly benefit from the existence of specially designed data structures, called indexes, that can help avoid exhaustively searching the entire database. However, existing indexing techniques for CDBMS like bitmaps, zonemaps etc. have been shown to result in large storage overhead and memory traffic between storage and CPU. Column imprints are an indexing approach for CDBMS that deals with these two issues by compressing the data in storage and by storing data in such a way that it reduces expensive data transfers between Random Access Memory (RAM) and Central Processing Unit (CPU) caches. However, data compression and decompression, which are necessary for query processing in CDBMS imposes a significant computational burden. To deal with this problem, parallel architectures, such as Graphical Processing Units (GPU) can be used. GPUs are co-processors that have been shown to outperform CPUs for highly parallel tasks. Besides this, GPUs are highly energy efficient, affordable, and available in many types of machines, from mobile devices to supercomputers. Despite their advantages, no work has focused on the use of GPUs for column imprints indexing. To address the gap mentioned before, this thesis introduces the first GPU algorithm, named GPUImprints, for indexing columnar databases with column imprints. We also performed the first experimental study, using one real-world dataset and one synthetic dataset, on the use of GPUs versus CPUs for column imprints in terms of their index creation time, query response times and index space requirements. Our experiments showed that GPUImprints can speedup the column imprints construction time by a factor of 20X and can speedup the query processing times of range queries by a factor of 3.7X when compared to the CPU imprints algorithm

Oliver Kullmann - One of the best experts on this subject based on the ideXlab platform.

  • Advanced Encryption Standard II benchmarks
    2013
    Co-Authors: Matthew Gwynne, Oliver Kullmann
    Abstract:

    [1], is a popular encryption cipher used in a variety of areas including wireless, disk, and network encryption. AES was designed to replace the Data Encryption Standard (DES), which is now vulnerable to brute force attacks, and to provide an alternative to the computationally expensive Triple-DES. Small-scale variants of the AES were introduced in [2]. Such small-scale variations allow the scaling of the AES block and key sizes while maintaining the cryptographic properties of the cipher. They were designed to offer instances which were more easily analysed, due to their smaller size, but which shared the same fundamental structure and features as the AES. We present benchmarks which are translations of smallscale AES into SAT. These benchmarks are generated via a general translation framework within the OKlibrary 1). Small-scale variants are used, instead of the standard AES, to provide instances which are likely to be solvable by SAT solvers within the time-frame of the SAT Challenge 2012. All work presented is available within the OKlibrary (see [3]), a research platform for hard problems; preliminary results were presented in [4], and a full exploration will be available in [5] (forthcoming). II. SMALL-SCALE AES A small-scale AES cipher aes(m, r, c, e) for m, r, c, e ∈ N, encrypts, by a series of m rounds. a plaintext P to ciphertext C using key K, where P, K and C are b-bit inputs and b: = r · c · e. Standard 128-bit AES corresponds to aes(10, 4, 4, 8) (with a small irrelevant change; see below). The b-bit inputs (P, K and C) constitute a matrix with r rows and c columns, and with elements in the byte field, a finite field of Order 2 e with base-set {0, 1} e (so “bytes ” have length e). These matrices are formed in Column-Major Order, that is, column-by-column, top-to-bottom, and left-to-right

Matthew Gwynne - One of the best experts on this subject based on the ideXlab platform.

  • Advanced Encryption Standard II benchmarks
    2013
    Co-Authors: Matthew Gwynne, Oliver Kullmann
    Abstract:

    [1], is a popular encryption cipher used in a variety of areas including wireless, disk, and network encryption. AES was designed to replace the Data Encryption Standard (DES), which is now vulnerable to brute force attacks, and to provide an alternative to the computationally expensive Triple-DES. Small-scale variants of the AES were introduced in [2]. Such small-scale variations allow the scaling of the AES block and key sizes while maintaining the cryptographic properties of the cipher. They were designed to offer instances which were more easily analysed, due to their smaller size, but which shared the same fundamental structure and features as the AES. We present benchmarks which are translations of smallscale AES into SAT. These benchmarks are generated via a general translation framework within the OKlibrary 1). Small-scale variants are used, instead of the standard AES, to provide instances which are likely to be solvable by SAT solvers within the time-frame of the SAT Challenge 2012. All work presented is available within the OKlibrary (see [3]), a research platform for hard problems; preliminary results were presented in [4], and a full exploration will be available in [5] (forthcoming). II. SMALL-SCALE AES A small-scale AES cipher aes(m, r, c, e) for m, r, c, e ∈ N, encrypts, by a series of m rounds. a plaintext P to ciphertext C using key K, where P, K and C are b-bit inputs and b: = r · c · e. Standard 128-bit AES corresponds to aes(10, 4, 4, 8) (with a small irrelevant change; see below). The b-bit inputs (P, K and C) constitute a matrix with r rows and c columns, and with elements in the byte field, a finite field of Order 2 e with base-set {0, 1} e (so “bytes ” have length e). These matrices are formed in Column-Major Order, that is, column-by-column, top-to-bottom, and left-to-right

Vani C Divya - One of the best experts on this subject based on the ideXlab platform.

  • VLSI IMPLEMENTATION OF HIGH PERFORMANCE MONTGOMERY MODULAR MULTIPLICATION FOR CRYPTO GRAPHICAL
    International Journal of Innovative Technology and Research, 2017
    Co-Authors: Vani C Divya
    Abstract:

    The multiplier factor receives and outputs the data with binary illustration and uses solely one-level Carry Save Adder (CSA) to avoid the carry propagation at each addition operation. This CSA is additionally accustomed perform operand pre-computation and format conversion from the carry save format to the binary illustration, leading to an occasional hardware price and short important path delay at the expense of additional clock cycles for finishing one standard multiplication. To beat the weakness, a Configurable CSA (CCSA), that may be one full-adder or 2 serial half-adders, is projected to scale back the additional clock cycles for quantity pre-computation and format conversion by 0.5. The mechanism which will notice and skip the surplus carry-save addition operations in the one-level CCSA design whereas maintaining the short important path delay is developed. The additional clock cycles for quantity pre-computation and format conversion is hidden and the high turnout is obtained. AES relies on a style principle called a substitution-permutation network, combination of each substitution and permutation, and is quick in each software package and hardware. AES doesn't use a Festal network. AES is a variant of Irondale that encompasses a fastened block size of 128 bits, and a key size of 128, 192, or 256 bits. By contrast, the Irondale specification in and of itself is nominative with block and key sizes that will be any multiple of thirty-two bits, both with a minimum of 128 and a most of 256 bits.AES operates on a 4×4 Column-Major Order matrix of bytes, termed the state, though some versions of Irondale has a bigger block size and have further columns within the state. Most AES calculations square measure tired a special finite field