RADIOELECTRONIC AND COMPUTER SYSTEMS

A concept of distributed replicated data storages like Cassandra, HBase, MongoDB has been proposed to effectively manage the Big Data sets whose volume, velocity, and variability are difficult to deal with by using the traditional Relational Database Management Systems. Trade-offs between consistency, availability, partition tolerance, and latency are intrinsic to such systems. Although relations between these properties have been previously identified by the well-known CAP theorem in qualitative terms, it is still necessary to quantify how different consistency and timeout settings affect system latency. The paper reports results of Cassandra's performance evaluation using the YCSB benchmark and experimentally demonstrates how to read latency depends on the consistency settings and the current database workload. These results clearly show that stronger data consistency increases system latency, which is in line with the qualitative implication of the CAP theorem. Moreover, Cassandra latency and its variation considerably depend on the system workload. The distributed nature of such a system does not always guarantee that the client receives a response from the database within a finite time. If this happens, it causes so-called timing failures when the response is received too late or is not received at all. In the paper, we also consider the role of the application timeout which is the fundamental part of all distributed fault tolerance mechanisms working over the Internet and used as the main error detection mechanism here. The role of the application timeout as the main determinant in the interplay between system availability and responsiveness is also examined in the paper. It is quantitatively shown how different timeout settings could affect system availability and the average servicing and waiting time. Although many modern distributed systems including Cassandra use static timeouts it was shown that the most promising approach is to set timeouts dynamically at run time to balance performance, availability and improve the efficiency of the fault-tolerance mechanisms.

timeout; NoSQL; distributed databases; replication; performance benchmarking; consistency; availability; latency; trade-off.

Meier, A., Kaufmann, M. SQL & NoSQL Data-bases: Models, Languages, Consistency Options and Architectures for Big Data Management, Berlin, Spring-er Verlag Publ., 2019. 229 p.

Pritchett, D. Base: An Acid Alternative. ACM Queue, 2008, vol. 6, no. 3, pp. 48-55.

Brewer, E. Towards Robust Distributed Systems. Proceedings of the 19th Ann. ACM Symp. on Principles of Distributed Computing, Portland, USA, 2000, pp. 7-8.

Gilbert, S., Lynch, N. Brewer’s Conjecture and the Feasibility of Consistent, Available, Partition-Tolerant Web Services. ACM SIGACT News, 2002, vol. 33, no. 2, pp. 51-59.

Cooper, B., Silberstein, A., Tam, E., Ramakrish-nan, R., Sears, R. Benchmarking Cloud Serving Systems with YCSB. Proceedings of the 1st ACM Symp. on Cloud Computing, Indianapolis, Indiana, USA, 2010, pp. 143-154.

Abramova, V., Bernardino, J., Furtado, P. Test-ing Cloud Benchmark Scalability with Cassandra. Pro-ceedings of the IEEE 10th World Congress on Services. Anchorage, USA, 2014, pp. 434-441.

Klein, J., Gorton, I., Ernst, N., Donohoe, P., Pham, K., Matser, C. Performance Evaluation of NoSQL Databases: A Case Study. Proceedings of the 1st ACM/SPEC Int. Workshop on Performance Analysis of Big Data Systems, Austin, USA, 2015, pp. 5-10.

Haughian, G., Osman, R., Knottenbelt, W. Benchmarking Replication in Cassandra and Mon-goDB NoSQL Datastores. Proceedings of the 27th Int. Conf. on Database and Expert Systems Applications, Porto, Portugal, 2016, pp. 152-166.

Farias, V. A., Sousa, F. R., Maia, J. G. R., Gomes, J. P. P., Machado, J. C. Regression based per-formance modeling and provisioning for NoSQL cloud databases. Future Generation Computer Systems, 2018, vol. 79, pp. 72-81.

Karniavoura, F. & Magoutis, K. A measure-ment-based approach to performance prediction in NoSQL systems. Proceedings of the 25th IEEE Int. Symp. on the Modeling, Analysis, and Simulation of Computer and Telecom. Systems, Banff, Canada, 2017, pp. 255-262.

Cruz, F., Maia, F., Matos, M., Oliveira, R., Pau-lo, J., Pereira, J., Vilaca, R. Resource usage prediction in distributed key-value datastores. Proceedings of the IFIP Distributed Applications and Interoperable Sys-tems Conf., Heraklion, Crete, 2017, pp. 144-159.

Abdelmoniem, A. M., Bensaou, B. Curbing Timeouts for TCP-Incast in Data Centers via A Cross-Layer Faster Recovery Mechanism. Proceedings of the IEEE Conf. on Computer Communications, Honolulu, HI, 2018, pp. 675-683.

Libman, L., Orda, A. Optimal retrial and timeout strategies for accessing network resources. IEEE/ACM Transactions on Networking, 2002, vol. 10, no. 4, pp. 551-564.

Avizienis, A., Laprie, J.-C., Randell, B., Land-wehr, C. Basic concepts and taxonomy of dependable and secure computing. IEEE Trans. on Dependable and Secure Computing, 2004, vol. 1, no. 1, pp. 11-33.

Brewer, E. CAP twelve years later: How the "rules" have changed. Computer, 2012, vol. 45, no. 2, pp. 23-29.

Gorbenko, A., Romanovsky, A., Tarasyuk, O. Fault tolerant internet computing: Benchmarking and modelling trade-offs between availability, latency and consistency. Journal of Network and Computer Appli-cations, 2019, vol. 146, pp. 1-14.

Gorbenko, A., Romanovsky, A. Time-outing Internet Services. IEEE Security & Privacy, 2013, vol. 11, no. 2, pp. 68-71.

Github, Benchmarking Cassandra and other NoSQL databases with YCSB. [Online]. Available at: https://github.com/cloudius-systems/osv/wiki/Benchmarking-Cassandra-and-other-NoSQL-databases-with-YCSB. (accessed 12.07.2020).

Gorbenko, A., Romanovsky, A., Tarasyuk, O. Interplaying Cassandra NoSQL consistency and per-formance: A benchmarking approach. In: S. Bernardi & e. all., eds. Communications in Computer and Infor-mation Science. Berlin: Springer Nature, 2020, pp. 168-184.

Gorbenko, A., Kharchenko, V., Tarasyuk, O., Chen, Y., Romanovsky, A. The threat of uncertainty in Service-Oriented Architecture. Proceedings of the RISE/EFTS Joint Int. Workshop on Software Engineer-ing for Resilient Systems, Newcastle, 2008, pp. 49-54.