A Real-Time Database Testbed and Performance Evaluation

Abstract

A lot of real-time database (RTDB) research has been done to process transactions in a timely fashion using fresh data reflecting the current real world status. However, most existing RTDB work is based on simulations. Due to the absence of a publicly available RTDB testbed, it is very hard to evaluate real-time data management techniques in a realistic environment. To address the problem, we design and develop an initial version of a RTDB testbed, called RTDB2 (Real-Time Database Benchmark), atop an open source database [5]. We develop soft real-time database workloads that model online stock trades, providing several knobs to specify workloads for RTDB performance evaluation. In addition, we develop a QoS management scheme in RTDB2 to detect overload and reduce workloads, via admission control and adaptive temporal data updates, under overload. From the extensive experiments using the stock trading workloads developed in RTDB2, we observe that adaptive updates can considerably improve the transaction timeliness. We also observe that admission control can only enhance the timeliness under severe overload, possibly causing underutilization problems for moderate workloads.

Chronos: Feedback Control of a Real Database System Performance

“…We have found considerable differences between simulation-based results and experimental results in a real database system [7]. The Chronos server processes data service requests from clients, while periodically pulling and updating 3,000 stock prices from Yahoo!…”

“…Very little prior work on real-time data management [1,9,14,7] has actually been implemented and evaluated in real database systems. However, no RTDB testbed is publicly available.…”

“…As a result, it is hard to perform RTDB research in a real database system. To address this problem, we have developed Chronos [7], which is significantly extended in this paper as discussed before.…”

“…To address the problem, we have designed and implemented a soft real-time database testbed, called Chronos [7], in which thousands of clients can send the database server requests for stock quotes, trades, and user portfolio updates. We have found considerable differences between simulation-based results and experimental results in a real database system [7].…”

Backlog Estimation and Management for Real-Time Data Services

“…Chronos provides four types of transactions: viewstock, view-portfolio, purchase, and sale for seven tables [9]. Our database schema and transactions are similar to TPC-W [22], which models a small number of tables and well-defined queries and transactions such as catalog browsing, purchase, and sale in an emulated online bookstore.…”