Publication 2001 Issue No. 8 - August Abstract - A Hybrid Solution of Fork/Join Synchronization in Parallel Queues
A Hybrid Solution of Fork/Join Synchronization in Parallel Queues
August 2001 (vol. 12 no. 8)
pp. 829-845
 ASCII Text x Ray Jinzhu Chen, "A Hybrid Solution of Fork/Join Synchronization in Parallel Queues," IEEE Transactions on Parallel and Distributed Systems, vol. 12, no. 8, pp. 829-845, August, 2001.
 BibTex x @article{ 10.1109/71.946659,author = {Ray Jinzhu Chen},title = {A Hybrid Solution of Fork/Join Synchronization in Parallel Queues},journal ={IEEE Transactions on Parallel and Distributed Systems},volume = {12},number = {8},issn = {1045-9219},year = {2001},pages = {829-845},doi = {http://doi.ieeecomputersociety.org/10.1109/71.946659},publisher = {IEEE Computer Society},address = {Los Alamitos, CA, USA},}
 RefWorks Procite/RefMan/Endnote x TY - JOURJO - IEEE Transactions on Parallel and Distributed SystemsTI - A Hybrid Solution of Fork/Join Synchronization in Parallel QueuesIS - 8SN - 1045-9219SP829EP845EPD - 829-845A1 - Ray Jinzhu Chen, PY - 2001KW - Fork/join queuing systemKW - modeling and analysisKW - dynamic-bubblesort analysisKW - upper boundsKW - parallel queuesKW - synchronization.VL - 12JA - IEEE Transactions on Parallel and Distributed SystemsER -

Abstract—A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of $K \geq 2$. The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate $\lambda$ and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, $k = 1, 2,\ldots, K$, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by $T_K$. We obtain a very simple (as a function of $T_1$ and $T_2$ only) and general upper bound expression for $T_K$ and we get an exact relationship between the cases for $K =2$ and 3. We evaluate our results by simulating $2, 3, \ldots, 99$, and 100 queues for $\rho = 0.1, 0.2,\ldots, 0.8$, and 0.9, each for four different HFJ cases, where $\rho =\lambda/\mu$ and $\mu$ is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent. Also, we compare our results with three recent papers [19], [20], [22].

[1] F. Baccelli, A. Makowski, and A. Shwartz, “The Fork-Join Queue and Related Systems with Synchronization Constraints: Stochastic Ordering, and Computable Bounds,” Advances in Applied Probability, vol. 21, pp. 629-660, Sept. 1989.
[2] F. Baccelli, W.A. Massey, and D. Towsley, "Acyclic Fork-Join Queueing Network", J. ACM, vol. 36, no. 3, July 1989.
[3] R.J. Chen, “Analysis and Simulation of Parallel Fork-Join System,” PhD dissertation, Dept. of Computer Science and Eng., Univ. of Connecticut, May 1998.
[4] R.J. Chen and H. Sholl, “100-Queue Simulation Report I for `Hybrid Solution of Fork/Join Synchronization in Parallel Queues',” Technique Report BRC/CSE-TR-2001, Univ. of Connecticut, 2000.
[5] S. Chen and D. Towsley, “Design and Modeling Policies for Two Server Fork/Join Queueing Systems,” COINS Technical Report 91-93, Univ. of Massachusetts, 1991.
[6] Y. Ding, L. Lipsky, and S. Kim, “On the Departure Process of a Fork-Join Queueing System,” Proc. ISMM Int'l Symp. Parallel and Distributed Computing, and Systems, Oct. 1990.
[7] L. Flatto and S. Hahn, “Two Parallel Queues Created by Arrivals with Two Demands I,” SIAM J. Applied Math., vol. 44, pp. 1041-1053, Oct. 1984.
[8] L. Flatto, “Two Parallel Queues Created by Arrivals with Two Demands II,” SIAM J. Applied Math., vol. 45, pp. 861-878, Oct. 1985.
[9] P. Heidelberger and K.S. Trivedi, “Queueing Network Models for Parallel Processing with Asynchronous Tasks,” IEEE Trans. Computers, vol. 31, no. 11, pp. 1099-1109, Nov. 1982.
[10] P. Heidelberger and K.S. Trivedi, “Analytic Queueing Models for Programs with Internal Concurrency,” IEEE Trans. Computers, vol. 32, no. 11, pp. 73-82, Nov. 1983.
[11] C. Kim and A.K. Agrawala, "Analysis of the Fork-Join Queue," IEEE Trans. Computers, vol. 38, pp. 250-255, no. 2, 1989.
[12] L. Kleinrock, Queueing Systems Vol. I: Theory. John Wiley and Sons, 1975.
[13] P. Konstantopoulos and J. Walrand, “Stationary and Stability of Fork-Join Networks,” J. Applied Probability, vol. 25, pp. 604-614, 1989.
[14] C.P. Kruskal and A. Weiss, "Allocating Independent Subtasks on Parallel Processors," IEEE Trans. Software Eng., vol. 11, no. 10, pp. 1,001-1,016, Oct. 1985.
[15] A. Kumar and R. Shorey, "Performance Analysis and Scheduling of Stochastic Fork-Join Jobs in a Mutilcomputer System," IEEE Trans. Parallel and Distributed. Systems, vol. 4, no. 10, pp. 1,147-1,164, Oct. 1993.
[16] L. Lipsky, Queueing Theorem, A Linear Algebraic Approach. New York: Macmillan, 1992.
[17] Y.C. Liu and H.G. Perros, “A Decomposition Procedure for the Analysis of a Closed Fork/Join Queueing System,” IEEE Trans. Computers, vol. 40, no. 3, pp. 365-370, Mar. 1991.
[18] R.R. Muntz and J.C.S. Lui, “Bounding the Response Time of a Minimum Expected Delay Routing System,” IEEE Trans. Computers, vol. 44, no. 5, pp. 1371-1382, Dec. 1995.
[19] J.C.S. Lui, R.R. Muntz, and D. Towsley, “Computing Performance Bounds of Fork-Join Parallel Programs under a Multiprocessing Environment,” IEEE Trans. Parallel and Distributed Systems, vol. 9, no. 3, pp. 295-311, Mar. 1998.
[20] S. Varma and A.M. Makowski, "Interpolation Approximations for Symmetric Fork-Join Queues," Proc. Performance '93, pp. 245-273,Rome, 1993.
[21] R.R. Muntz, E. De Souza, E Silva, and A. Goyal, "Bounding Availability of Repairable Computer Systems," Proc. 1989 ACM SIGMETRICS and PERFORMANCE '89. Also in special issue of IEEE Trans. Computers, vol. 38, no. 12, pp. 19-30, Dec. 1989.
[22] R. Nelson and A. Tantawi, “Approximate Analysis of Fork/Join Synchronization in Parallel Queues,” IEEE Trans. Computers, vol. 37, pp. 739–743, June 1988.
[23] R. Nelson, D. Towsley, and A. Tantawi, “Performance Analysis of Parallel Processing Systems,” IEEE Trans. Software Engineering, vol. 14, pp. 532–539, Apr. 1988.
[24] H. Sholl, R.J. Chen, and J. Moriarty, “Real-Time N-Queue Dynamic Schedule for M/G/1 System,” Technical Report BRC/CSE-TR-94-12, Univ. of Connecticut, Storrs, pp. 1-14, 1992.
[25] A. Thomasian and A. Tantawi, "Approximate Solutions for M/G/1 Fork-Join Synchronization," Proc. Winter Simulation Conf., pp. 361-368,Orlando, Fla., Dec. 1994.

Index Terms:
Fork/join queuing system, modeling and analysis, dynamic-bubblesort analysis, upper bounds, parallel queues, synchronization.
Citation:
Ray Jinzhu Chen, "A Hybrid Solution of Fork/Join Synchronization in Parallel Queues," IEEE Transactions on Parallel and Distributed Systems, vol. 12, no. 8, pp. 829-845, Aug. 2001, doi:10.1109/71.946659