C. Battista et al.: `The APE-100 Computer: (I) the Architecture', Int. J. of High Speed Computing 5 (1993) 637.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....facilities, exploiting nearest neighbor connectivity. The novel method is based on the hyper systolic algorithm. We discuss its implementation on APE100 Quadrics and present bandwidth measurements. 1 Introduction Massively parallel SIMD systems like the Italian supercomputer APE100 Quadrics[1] are very attractive for high performance computing applications as they offer a pathway to Teracomputing in the near future[2] at an affordable price. The 3 dimensional interprocessor network of the supercomputer APE100 Quadrics has been designed to carry out fast data movements in quantum ....

C. Battista et al.: `The APE-100 Computer: (I) the Architecture', Int. J. of High Speed Computing 5 (1993) 637.

....leading to a scalable implementation [18] 3. Cache or register access optimization can be organized as in conventional methods. 4 Implementation and Results 4. 1 The APE100 Quadrics Parallel Computer APE100 Quadrics is a parallel computer which has been designed by Italian physicists at INFN [19] for simulations in the field of Quantum Chromodynamics. 0 0 0) 1 0 0) 1 1 0) 0 1 0) 0 1 1) 0 0 1) 1 0 1) 1 1 1) Fig. 2. Geometry of the 8 node board of the APE100 Quadrics. Larger assemblies of basic elements can be put together where the network is toroidally connected in each ....

C. Battista et al.: `The APE-100 Computer: (I) the Architecture', Int. J. of High Speed Computing 5 (1993) 637. This article was processed using the L A T E X macro package with LLNCS style

....intensive but, on the other hand, it is regular and thus well suited for cost effective SIMD systems, phase (ii) is more involved but less compute intensive and thus can be advantageously implemented on general purpose MIMD systems. The SIMD platforms we exploit are the parallel APE100 systems [6, 7], developed by Italian physicists at INFN as very effective lattice QCD engines. APE100 machines are SIMD parallel computers which are based on a regular 3 dimensional grid connectivity. They have gained attention in a variety of fields which require the power of parallel supercomputers. APE100 ....

.... APE100 machines has been considerably enlarged by use of systolic and hyper systolic algorithms for linear algebra tasks and FFT [19 25] The APE100 Quadrics parallel computer has been designed for quantum field theory, mainly as cheap compute engine in the field of lattice quantum chromodynamics [6, 7, 26, 27]. Therefore, its endowments are reduced and it is optimized for efficient multiplications of 3 Theta 3 matrices and 3 vectors. Diagram Fig. 1 depicts the Quadrics system: Compilation and program submission as FRONT END Z CPU COMPUTE NODES (0 0 , 0) 1 0 0) 1 1 0) 0 1 0) 0 1 1) 0 1) 1 0 1) ....

C. Battista et al.: `The APE-100 Computer: (I) the Architecture', Int. J. of High Speed Computing 5 (1993) 637.

No context found.

Sez. Roma. [2] C. Battista et al. " The APE-100 Computer: (I) The Architecture" Int. Journ. High Speed Comp. 5 (1993) 637-656; A. Bartoloni et al. "A Hardware Implementation of the APE100 Architecture" Int.

No context found.

C. Battista et al.: `The APE-100 Computer: (I) the Architecture', Int. J. of High Speed Computing 5 (1993) 637.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC