TY - GEN
T1 - The effect of workload on the performance and availability of voting algorithms
AU - Qureshi, M. A.
AU - Sanders, W. H.
PY - 1995
Y1 - 1995
N2 - Voting algorithms are a popular way to provide data consistency in replicated data systems. By maintaining multiple copies of data on distinct servers, they can increase the data's availability, as perceived by a user. Many models have been made to study the degree to which replication increases the availability of data, and some have been made to study the cost incurred in maintaining consistency. However, little work has been done to evaluate the time it takes to serve a request, accounting for server and network failures, or to determine the effect of workload on these measures. The effect of workload can be significant, since failures of system components are not important unless they are needed to deliver a service, and requests can force updates on data that would otherwise be outdated. In this paper, with the help of stochastic activity networks, we determine the availability and mean time to respond to write requests as a function of the number of replicated copies and workload offered to the system. The results illustrate that it is indeed possible to determine such measures analytically and that workload, as well as the number of copies, is an important determinant of availability and response time.
AB - Voting algorithms are a popular way to provide data consistency in replicated data systems. By maintaining multiple copies of data on distinct servers, they can increase the data's availability, as perceived by a user. Many models have been made to study the degree to which replication increases the availability of data, and some have been made to study the cost incurred in maintaining consistency. However, little work has been done to evaluate the time it takes to serve a request, accounting for server and network failures, or to determine the effect of workload on these measures. The effect of workload can be significant, since failures of system components are not important unless they are needed to deliver a service, and requests can force updates on data that would otherwise be outdated. In this paper, with the help of stochastic activity networks, we determine the availability and mean time to respond to write requests as a function of the number of replicated copies and workload offered to the system. The results illustrate that it is indeed possible to determine such measures analytically and that workload, as well as the number of copies, is an important determinant of availability and response time.
KW - Availability
KW - Replicated Data Systems
KW - Response Time
KW - Stochastic Activity Networks
KW - Stochastic Petri Nets
KW - Voting Algorithms
UR - http://www.scopus.com/inward/record.url?scp=2442462692&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=2442462692&partnerID=8YFLogxK
U2 - 10.1109/MASCOT.1995.378684
DO - 10.1109/MASCOT.1995.378684
M3 - Conference contribution
AN - SCOPUS:2442462692
T3 - Proceedings - IEEE Computer Society's Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems, MASCOTS
SP - 217
EP - 224
BT - Proceedings of the 3rd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 1995
PB - IEEE Computer Society
T2 - 3rd International Workshop on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 1995
Y2 - 18 January 1995 through 20 January 1995
ER -