Abstract: In-memory database systems are becoming popular
due to the availability and affordability of sufficiently large RAM and
processors in modern high-end servers with the capacity to manage
large in-memory database transactions. While fast and reliable inmemory
systems are still being developed to overcome cache misses,
CPU/IO bottlenecks and distributed transaction costs, disk-based data
stores still serve as the primary persistence. In addition, with the
recent growth in multi-tenancy cloud applications and associated
security concerns, many organisations consider the trade-offs and
continue to require fast and reliable transaction processing of diskbased
database systems as an available choice. For these
organizations, the only way of increasing throughput is by improving
the performance of disk-based concurrency control. This warrants a
hybrid database system with the ability to selectively apply an
enhanced disk-based data management within the context of inmemory
systems that would help improve overall throughput.
The general view is that in-memory systems substantially
outperform disk-based systems. We question this assumption and
examine how a modified variation of access invariance that we call
enhanced memory access, (EMA) can be used to allow very high
levels of concurrency in the pre-fetching of data in disk-based
systems. We demonstrate how this prefetching in disk-based systems
can yield close to in-memory performance, which paves the way for
improved hybrid database systems. This paper proposes a novel EMA
technique and presents a comparative study between disk-based EMA
systems and in-memory systems running on hardware configurations
of equivalent power in terms of the number of processors and their
speeds. The results of the experiments conducted clearly substantiate
that when used in conjunction with all concurrency control
mechanisms, EMA can increase the throughput of disk-based systems
to levels quite close to those achieved by in-memory system. The
promising results of this work show that enhanced disk-based
systems facilitate in improving hybrid data management within the
broader context of in-memory systems.
Abstract: The speculative locking (SL) protocol extends the twophase locking (2PL) protocol to allow for parallelism among conflicting transactions. The adaptive speculative locking (ASL) protocol provided further enhancements and outperformed SL protocols under most conditions. Neither of these protocols consider the impact of network latency on the performance of the distributed database systems. We have studied the performance of ASL protocol taking into account the communication overhead. The results indicate that though system load can counter network latency, it can still become a bottleneck in many situations. The impact of latency on performance depends on many factors including the system resources. A flexible discrete event simulator was used as the testbed for this study.
Abstract: A new approach for timestamp ordering problem in
serializable schedules is presented. Since the number of users using
databases is increasing rapidly, the accuracy and needing high
throughput are main topics in database area. Strict 2PL does not
allow all possible serializable schedules and so does not result high
throughput. The main advantages of the approach are the ability to
enforce the execution of transaction to be recoverable and the high
achievable performance of concurrent execution in central databases.
Comparing to Strict 2PL, the general structure of the algorithm is
simple, free deadlock, and allows executing all possible serializable
schedules which results high throughput. Various examples which
include different orders of database operations are discussed.
Abstract: Designing, implementing, and debugging concurrency
control algorithms in a real system is a complex, tedious, and errorprone
process. Further, understanding concurrency control
algorithms and distributed computations is itself a difficult task.
Visualization can help with both of these problems. Thus, we have
developed an exploratory environment in which people can prototype
and test various versions of concurrency control algorithms, study
and debug distributed computations, and view performance statistics
of distributed systems. In this paper, we describe the exploratory
environment and show how it can be used to explore concurrency
control algorithms for the interactive steering of distributed
computations.