Latin cubes and parallel array access

Fan, Cong; Gupta, Ajay; Liu, Jiuquang

doi:10.1109/ipps.1994.288310

Cited by 4 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [40] , constructions have been presented for perfect latin squares of order N that are used in skewing schemes that permit conflict-free access to rows , columns , main diagonals and main subsquares of the matrix A . The idea is extended further in [42] , to three-dimensional data structures , where an optimal skewing-scheme is implemented using the three-dimensional counterparts of latin squares , viz . latin cubes .…”

Section: Iia Parallel Array Accessmentioning

confidence: 99%

Combinatorial Techniques and Objects in Computer Science: Fault-tolerance and Other Interesting Applications

Singhi

Shriniwas²,

Antony³

1996

European Journal of Combinatorics

View full text Add to dashboard Cite

The main objective of this contribution is to outline briefly , a set of interesting problems in computer science that involve the use of combinatorial objects and techniques . We also present our results for the case of fault-tolerance in hypercube-based parallel computer architectures , where we show the involvement of t -designs and t -covers in the structure of extremal fault-sets , which is useful in the construction of fault-tolerant m -partitions of the n -cube .÷ 1996 Academic Press Limited I NTRODUCTION In recent years , many complex problems have been encountered en route to the development of new technologies and methodologies responsible for the advances made in computer science . Some of these problems are presently being tackled using existing results from combinatorics ; in certain cases , the requirements of the computer science problems have also led to the refinement of existing combinatorial techniques , and even to the development of new results in combinatorics . We present a set of examples from current research areas in computer science , and briefly outline the usage of combinatorial objects such as t -designs , latin squares , latin cubes , SBIBDs , simplicial complexes and so on . This paper also illustrates the use of traditional combinatorial objects such as codes for of f-beat applications that focus more on the inherent properties of elements of a code , rather than just error-detection / correction abilities . This paper is organized as follows : in Part A we give examples from the areas of fault-tolerance (Section I) , parallel processing (Section II) , computer communications and networking (Section III) and online transaction processing (Section IV) ; in Part B , we briefly present our results in the areas of fault-tolerance . Part A I . F AULT -T OLERANCE Ia . Problem : codes for memory subsystemsDue to advances in the technology of construction of memory and related storage subsystems (such as RAM , ROM , CD-ROM , WORM etc . ) , most of the errors that occur in the binary stored information are of the unidirectional type . Unidirectional errors within a storage unit are caused by noise , faulty recording media etc . and as a result either some 0 bits in the information get converted to 1 bits or vice versa . The percentage of truly random errors , in which both types of errors occur , is very small in proportion to the unidirectional errors , due to the advanced recording techniques employed . For instance , in the case of optical media (such as CD-ROMs) only the 1 -0 type of error is possible ; the 0 -1 type is not possible . Thus , any modern error detecting / correcting code must be capable of handling a large number of unidirectional errors in addition to handling a relatively small number of truly symmetric random errors . Such codes are called t -error-correcting , k -error-detecting and d -unidirectional-97 0195-6698 / 96 / 020097 ϩ 15 $18 . 00 / 0 ÷ 1996 Academic Press Limited N . M . Singhi et al . 98error-detecting codes , with d Ͼ k Ͼ t . There are also t -unidirecti...

show abstract

Section: Iia Parallel Array Accessmentioning

confidence: 99%

Combinatorial Techniques and Objects in Computer Science: Fault-tolerance and Other Interesting Applications

Singhi

Shriniwas²,

Antony³

1996

European Journal of Combinatorics

View full text Add to dashboard Cite

show abstract

Data space mapping for efficient I/O in large multi-dimensional databases

Ferhatosmanoğlu

Ramachandran

Agrawal

et al. 2007

Information Systems

View full text Add to dashboard Cite

A lightweight I/O scheme to facilitate spatial and temporal queries of scientific data analytics

Tian

Liu

Klasky

et al. 2013

2013 IEEE 29th Symposium on Mass Storage Systems and Technologies (MSST)

View full text Add to dashboard Cite

In the era of petascale computing, more scientific applications are being deployed on leadership scale computing platforms to enhance the scientific productivity. Many I/O techniques have been designed to address the growing I/O bottleneck on large-scale systems by handling massive scientific data in a holistic manner. While such techniques have been leveraged in a wide range of applications, they have not been shown as adequate for many mission critical applications, particularly in data postprocessing stage. One of the examples is that some scientific applications generate datasets composed of a vast amount of small data elements that are organized along many spatial and temporal dimensions but require sophisticated data analytics on one or more dimensions. Including such dimensional knowledge into data organization can be beneficial to the efficiency of data post-processing, which is often missing from exiting I/O techniques. In this study, we propose a novel I/O scheme named STAR (Spatial and Temporal AggRegation) to enable high performance data queries for scientific analytics. STAR is able to dive into the massive data, identify the spatial and temporal relationships among data variables, and accordingly organize them into an optimized multi-dimensional data structure before storing to the storage. This technique not only facilitates the common access patterns of data analytics, but also further reduces the application turnaround time. In particular, STAR is able to enable efficient data queries along the time dimension, a practice common in scientific analytics but not yet supported by existing I/O techniques. In our case study with a critical climate modeling application GEOS-5, the experimental results on Jaguar supercomputer demonstrate an improvement up to 73 times for the read performance compared to the original I/O method.

show abstract

Latin cubes and parallel array access

Cited by 4 publications

References 11 publications

Combinatorial Techniques and Objects in Computer Science: Fault-tolerance and Other Interesting Applications

Combinatorial Techniques and Objects in Computer Science: Fault-tolerance and Other Interesting Applications

Data space mapping for efficient I/O in large multi-dimensional databases

A lightweight I/O scheme to facilitate spatial and temporal queries of scientific data analytics

Contact Info

Product

Resources

About