Many recent studies have suggested that the optimistic concurrency control (OCC) protocols outperform the locking-based protocols in real-time database systems (RTDBS). However, the OCC protocols suffer from the problem of unnecessary transaction restarts that is detrimental to transactions meeting their deadlines. The problem is more intensified in mixed transaction environments. Firm transactions are more vulnerable to restarts when they are in conflict with hard transactions on data access. In this paper, we addressed the problem and devised an effective OCC protocol with dynamic adjustment of serialization order, called OCC-DA, for RTDBS with mixed transactions. This protocol can avoid unnecessary transaction restarts by dynamically adjusting the serialization order of the conflicting transactions with respect to the validating transaction. As a result, much resource can be saved and more firm transactions can meet their deadlines without affecting the execution of hard transactions. The characteristics of the OCC-DA protocol were examined in detail by simulation. The results show that the performance of the OCC-DA protocol was consistently better than the other two popular protocols, OCC with forward validation and OCC with Wait-50, over a wide range of system settings. In particular, the OCC-DA protocol provides a more significant performance gain in mixed transaction environments.

In this paper, we investigate the approach of using separate algorithms to process read-only transactions in real-time systems. A read-only transaction (ROT) is a transaction that only reads, but does not update any data item. Since there is a significant proportion of ROTs in several real-time systems, it is important to investigate how to process ROTs effectively. Using an algorithm to process ROTs separately from update transactions may reduce the interference between ROTs and update transactions. This reduced interference alleviates the impact of concurrency control on real-time priority-driven scheduling and improves the timeliness of the system. Moreover, we explore the different consistency requirements of ROTs. Particularly, we define a weaker form of consistency, view consistency, which allows ROTs to perceive different serialization order of update transactions. While view consistency permits non-serializability, ROTs are still ensured to see consistent data. We propose two robust algorithms for different consistency requirements of ROTs. The two algorithms are robust in the sense that they can be used in a compatible way so that a real-time system can provide different consistent data for different applications. The performance of two algorithms was examined through a series of simulation studies. The simulation results show that the two algorithms outperform the high-priority two-phase locking protocol.

The difficulties of providing a guarantee of meeting transaction deadlines in hard real-time database systems lie in the problems of priority inversion and of deadlocks. Priority inversion and deadlock problems ensue when concurrency control protocols are adapted in priority-driven scheduling. The blocking delay due to priority inversion can be unbounded, which is unacceptable in the mission-critical real-time applications. Some priority ceiling protocols have been proposed to tackle these two problems. However, they are too conservative in scheduling transactions for the single-blocking and deadlock-free properties, leading to many unnecessary transaction blockings. The authors analyze the unnecessary transaction blocking problem inherent in these priority ceiling protocols and investigate the conditions for allowing a higher priority transaction to preempt a lower priority transaction using the notion of dynamic adjustment of serialization order. A new priority ceiling protocol is proposed to solve the unnecessary blocking problem, thus enhancing schedulability. They also devise the worst-case schedulability analysis for the new protocol which provides a better schedulability condition than other protocols.

This paper presents a performance study of ATM networks in the support of real-time MPEG-I video transmission. Multiple classes of MPEG-I are used in the study and what makes this simulation study different from the others is the video data used. These data are captured from real video programmes and we categorized these video clips according to their workload characteristics. The performance of the ATM switch is examined in terms of the cell miss ratio due to deadline missing and the cell loss ratio due to buffer overflow in both the ATM switch as well as the gateways. Moreover, a higher level of abstraction in terms of burst loss ratio is also collected. The results indicate that the first-come-first-serve scheduling algorithm is insufficient to handle real-time traffic. This paper presents a better method to improve the performance significantly especially when the virtual path bandwidth negotiated is conservative.

Recently the concurrency control issue of real-time transactions is gaining increasing attention of researchers in the database community. One of the major design issue in concurrency control of real-time transactions is the resolution of local as well as distributed deadlocks while at the same time meeting the timing requirements of the transactions. In this paper, a new deadlock detection algorithm specially designed for distributed real-time database systems is proposed. The performance of the proposed algorithm is evaluated through extensive simulation experiments. Studies have also been carried out to compare the performance of the real-time deadlock detection algorithm with a non real-time algorithm for both firm and soft real-time transactions. Results indicated that the real-time deadlock detection algorithm performs better than the non real-lime deadlock detection algorithm. Results also indicated that the performance of the new algorithm is substantially better for soft real-time than that of firm real-time systems.

The design of concurrency control protocols for time-constrained transactions is complicated due to the requirements to maintain the database consistency and to satisfy the timing constraints of the transactions. In the past few years, various real-time locking protocols have been proposed for different real-time database systems (RTDBS). However, the use of these protocols for distributed real-time database (DRTDBS) has received much less attention, even though many RTDBS are distributed in nature. In this paper, two efficient real-time locking protocols are proposed for DRTDBS. The first one, based on dynamic locking, is called Distributed Hybrid Two Phase Locking (DHb2PL). Its performance has been compared in detail with three other distributed real-time locking protocols. The performance results indicate that DHb2PL is much better than the other protocols as a result of a better approach to resolving lock conflicts and its deadlock free property. The second one, called DRT-S2PL, is based on static locking where the locks required by a transaction are assumed to be known before its execution. Its relative performance as compared with DHb2PL is dependent on the proportion of remote locks required by a transaction. DRT-S2PL is more suitable for systems with transactions which have to set a large proportion of remote locks.