PARALLEL CALIBRATED EMULATION AS A TECHNIQUE FOR EVALUATING PARALLEL ARCHITECTURES

Citation
Hl. Muller et al., PARALLEL CALIBRATED EMULATION AS A TECHNIQUE FOR EVALUATING PARALLEL ARCHITECTURES, Computer systems science and engineering, 13(1), 1998, pp. 17-25
Citations number
16
Categorie Soggetti
Computer Science Hardware & Architecture","Computer Science Theory & Methods","Computer Science Theory & Methods","Computer Science Hardware & Architecture
ISSN journal
02676192
Volume
13
Issue
1
Year of publication
1998
Pages
17 - 25
Database
ISI
SICI code
0267-6192(1998)13:1<17:PCEAAT>2.0.ZU;2-#
Abstract
We describe the use of a calibrated emulator to simulate a parallel co mputer architecture. The emulator has a virtual clock, but unlike the virtual clock of a simulator, the emulator clock is bound to a fixed f raction of real time. Individual processors time actions independently , thus without the need for a globally synchronised clock value. Each component of the emulator is calibrated (by slowing it down artificial ly) so that the balance of the speeds of all components reflects the b alance of the system under consideration. Unlike an ordinary simulator , a calibrated emulator is inherently parallel. The technique has been applied in the form of a parallel transputer-based emulator developed to evaluate the DDM - a scalable virtual shared memory architecture. The emulator provides performance results of a hardware implementation of the DDM using a calibrated virtual clock. A large transputer platf orm is used to run experiments. A couple of hours are sufficient to em ulate the execution of a realistic application on a large DDM.