Transaction Failure Research Articles

On handling component and transaction failures in multi agent systems

Multi agent systems are being used for various practical applications like e-commerce, e-auctions and gathering information from the web. Thus there is a need for these systems to be robust. However, agents can fail due to component failures. The atomic tasks taken up by the agents might also fail. So, agents need to recover to a correct state after a failure. This paper deals with the logging required and hence forth the recovery protocol to recover the agent to a correct state. An agent is modeled as a colored Petrinet. By logging the state information of the agent (Petrinet) an agent recovers from failures. For atomicity of tasks in multi-agent systems, linear two-phase commit protocol with some enhancements is proposed

The M/G/oo Queueing Model for Optimistic Concurrency Control

The M/G/∞ queue with Optimistic Concurrency Control (OCC) is a model for a special form of parallel transaction processing in a real-time database. Transactions arrive according to a Poisson process and require some generally distributed execution time. One of the differences with ordinary multi- and infinite-server queueing models is that under OCC the successful completion of one transaction may immediately cause the failure of one or more of the other transactions. This happens if the completing transaction has overwritten a data-item that is in use by another transaction in progress. As soon as this failure is detected the failed transaction is restarted. So the total service time of a transaction consists of its final successful run and the time spent on unsuccessful runs. In this study we develop an approximation for the distribution of the total service time, and test the approximation against simulation. Although in practice the number of servers is never unlimited, this study provides valuable insight in the asymptotics with respect to the number of servers. The approximation clearly demonstrates the very limited performance gain from an increase of the number of servers.

Resolving Executing-Committing Conflicts in Distributed Real-time Database Systems

In a distributed real-time database system (DRTDBS), a commit protocol is required to ensure transaction failure atomicity. If data conflicts occur between executing and committing transactions, the performance of the system may be greatly affected. In this paper, we propose a new protocol, called deadline-driven conflict resolution (DDCR), which integrates concurrency control and transaction commitment management for resolving executing and committing data conflicts amongst firm real-time transactions. With the DDCR, a higher degree of concurrency can be achieved, as many data conflicts of such kind can be alleviated, and executing transactions can access data items which are being held by committing transactions in conflicting modes. Also, the impact of temporary failures which occurred during the commitment of a transaction on other transactions, and the dependencies created due to sharing of data items is much reduced by reversing the dependencies between the transactions. A simulation model has been developed and extensive simulation experiments have been performed to compare the performance of the DDCR with other protocols such as the Opt [1], the Healthy-Opt [2], and the base protocol, which use priority inheritance and blocking to resolve the data conflicts. The simulation results show that the DDCR can significantly improve the system performance under different workload and workload distributions. Its performance is consistently better than the base protocol and the Opt protocols in both main-memory resident and disk-resident DRTDBS.

Restitution–Mistake of Law–Reform in Haste, Repent at Leisure

In Kleinwort Benson v. Lincoln City Council [1998] 3 W.L.R. 1095 the House of Lords heard consolidated appeals arising out of the failure of interest rate swap transactions entered into between various banks and local authorities. Having paid out considerable sums under the swap arrangements, the banks were disconcerted to find them declared ultra vires and void (Hazell v. Hammersmith and Fulham L.B.C. [1992] 2 A.C. 1). The Lords have now held that sums paid by the banks may be recovered back, by action in restitution for mistake. In so holding, they have abrogated the rule that a mistake of law is not actionable, and denied (at least on these facts) any defence that the parties acted on the basis of “settled law”.

Some constraints assuring serializability in multidatabases

In a multidatabase system, global transactions are executed under the control of the global system, while local transactions are submitted to each local system being outside of the global control. Since the multidatabase system is not aware of the local transactions and the conflicts they may cause, it has difficulty in detecting such conflicts. In [1], the authors gave the following constraints: global transactions should be commit-deferred and local schedules rigorous to assure global serializability of the multidatabase system. In this paper, we define dynamic-serializability and some classes of safe schedules which are recoverable from failure of transactions. Using them we give different, something weaker constraints, while global serializability of the multidatabase system is still achieved.

Mobile architecture for the Wishard Memorial Hospital Ambulatory Service

The introduction of portable computers into existing database systems requires traditional transaction mechanisms to be adapted to meet the needs of the mobile user. These mobile users utilise hand-held and laptop computers called mobile hosts, that can access the database from any location at any time. In this paper, we describe a mobile architecture that places mobile on the ambulances of the Wishard Memorial Hospital Ambulatory Service. This mobile architecture is designed to support transactions created by a mobile host that may be disconnected from the network during the execution of these transactions. The environment for this mobile architecture is the Indianapolis Network for Patient Care and Research (INPCR) of Marion County, Indianapolis, Indiana. The INPCR, which was developed to provide patient care by presenting clinical data in a clear and efficient manner, utilises the distributed medical patient database of the Regenstrief Medical Record System (RMRS) as a data repository. The medical personnel of the ambulatory service are able to submit transactions to the patient database of the RMRS to access vital patient information. The transactions in this environment use the forward recovery of sagas to minimise the amount of recovery needed during transaction failure. By allowing the medical personnel (hospital emergency room staff and the emergency medical technicians of the Wishard Memorial Hospital Ambulance Service) the ability to store, retrieve, create, and evaluate the medical records of their patients, they can provide high-quality care in a time and cost effective manner.

An extended transaction to maintain consistency and recovery in multidatabase systems

A complete practical solution to transaction management preserving multidatabase consistency in the presence of multidatabase updates and failures is presented. The approach developed does not require support for the two-phase commit (2PC) protocol in the participating local database management systems (LDBMSs). Furthermore, it does not violate local autonomy; the source code of the LDBMSs is not modified in any way and the multidatabase system (MDBS) does not access or modify any control information of the LDBMS. The principles of the 2PC protocol in the process of global transaction commitment are adopted. The presented method does not rely on any specific concurrency control mechanism for LDBMSs. Consideration is given to global transaction failures due to subtransaction aborts by the LDBMSs and local site crashes. The recovery process is based on undo operations. While a global transaction is in progress, the tables accessed by subtransactions of this transaction at each local site are locked using specially initiated table locks. These locks are stored and maintained in the local database itself as control tables. The approach taken is easy to implement, and its limitations are discussed.

Why franchise companies expand overseas

Business format franchising is becoming an increasingly international activity. From 1971 to 1985, U.S. franchisors added foreign outlets at a rate of 17% per year, almost twice as fast as they added domestic outlets (Aydin and Kacker 1990). As a result, by 1990 more than 350 U.S. companies had more than 32,000 franchised outlets overseas. By 2000, 60% of all franchisors in the United States are expected to have outlets overseas (Hoffman and Preble 1993). This study examines the 815 largest U.S. franchisors to understand what capabilities encourage them to expand overseas. It finds that the key capability that predicts the intent to expand overseas is superior capability to reduce franchisee opportunism. Franchisors who seek foreign franchisees have developed a greater capability to bond against and monitor potential franchisee opportunism. The data show that these differences are consistent across all industries in which franchising takes place. The results of this study indicate that foreign entrepreneurs can identify the American franchisors most likely to expand overseas by looking at their pricing structure and their monitoring capabilities. The easy identification of characteristics from which to find American franchisors will help to reduce the search costs of potential foreign franchisees. This reduction in search costs will make the establishment of international franchise relationships less expensive. This study also provides guidance to franchisors interested in expanding overseas. The results show how franchisors can structure their franchise relationships to reduce potential franchisee opportunism. This ability to reduce franchisee opportunism will make it easier for franchisors to enter high-growth foreign markets using the franchising business mode. This study also has implications for researchers. It suggests that international business research examine further the mechanisms by which firms make contractual modes of international business work. Whereas many firms may internalize international market transactions under conditions likely to lead to market failure, the large number of franchisors who use franchising as an international expansion mode despite conditions of market failure suggests that more attention be paid to mechanisms that companies can use to reduce the probability of failure of international contractual transactions. By helping to explain how franchisors monitor foreign franchisees or bond them against opportunistic behavior, this study suggests that the international business literature develop a more complex understanding of the workings of international business transactions than the simple choice of internalization or contractual entry modes.

The Effect of Failures on the Performance of Long-Duration Database Transactions

The occurrence of transaction failures is often unavoidable in most database processing environments and such failures can have a particularly detrimental effect on long-duration transactions. In this paper, the impact of failures on the performance of long-duration database transactions is studied, and results for the transaction response time and failure overhead are presented for general failure patterns, which include Poisson failures as a special case. It is found that the failure overhead can be excessive for long-duration transactions and the transaction response time can increase sharply with increased failure rates. The benefits of transaction nesting to arrest performance deterioration is considered and results for quantifying the extent of restructuring improvement are provided. A procedure based on branch and bound techniques for determining the optimal restructuring strategy to minimize total transaction response time is also presented. It is shown that restructuring optimization can bring about significant performance advantages.

Quorum consensus in nested-transaction systems

Gifford's Quorum Consensus algorithm for data replication is studied in the context of nested transactions and transaction failures (aborts), and a fully developed reconfiguration strategy is presented. A formal description of the algorithm is presented using the Input/Output automaton model for nested-transaction systems due to Lynch and Merritt. In this description, the algorithm itself is described in terms of nested transactions. The formal description is used to construct a complete proof of correctness that uses standard assertional techniques, is based on a natural correctness condition, and takes advantage of modularity that arises from describing the algorithm as nested transactions. The proof is accomplished hierarchically, showing that a fully replicated reconfigurable system “simulates” an intermediate replicated system, and that the intermediate system simulates an unreplicated system. The presentation and proof treat issues of data replication entirely separately from issues of concurrency control and recovery.

Transaction recovery in federated autonomous databases

Transaction management for federated database systems (FDBSs), where each participating DBMS is autonomous, supports global transactions that can access more than one database. A number of concurrency control algorithms exist for the failure-free environment. Handling transaction failure, due to concurrency control or site related reasons, becomes especially difficult in the presence of autonomy. Due to local autonomy, carrying out 2-phase commit may not be possible. This can be simulated by providing a server on top of the pre-existing DBMS at each site, which is responsible for submitting the local operations to the associated DBMS and communicating with the transaction's originating site. In this paper we formalize the problem of ensuring transaction consistency in an FDBS environment in the presence of failure. The key problem is that due to autonomy, the local DBMS and FDBS may have different views of an execution sequence generated at a site.Local recoverability is identified as the property of local execution sequences necessary for correctness. The other main problem is of guaranteeing that the various locally recoverable histories are consistent with each other. These identified properties are necessary and sufficient conditions for the correctness of FDBS recovery algorithms, and can be used to evaluate the correctness of the proposed algorithms. This paper also presents an FDBS recovery algorithm that has been proved to be correct. Formal proofs of all properties and a comparison of different algorithms are provided.

Starting up in a turbulent environment: A process model of failure among firms with high customer dependence

This paper proposes a process model on how new and small organizations may fail. Based on prior research, the paper builds a foundation for a process model, and then elaborates it inductively by explaining how and why significant reversals and shifts occurred between 1983 and 1985 in the entrepreneurial, organizational, and ecological conditions of ten new educational software companies. The process model states that all organizations engage in transactions with others in the normal conduct of business, and it is inevitable that a certain proportion of these transactions fail. The probability of transaction failures is a product of the failure rate at the level of the organizational population, as well as the kinds of transactions in which individual firms engage. New and small firms in a turbulent environment represent a subset of the organizational population particularly vulnerable to the chance occurrence of a failed transaction. Because of liabilities of newness and small size, these firms lack legitimacy. Furthermore, in many industries there is an absence of tangible physical assets that entrepreneurs can use as collateral to attract valuable resources and customers. In such situations, entrepreneurs often use transactions with key external constituencies as a surrogate to attract other resource constituents, thus giving rise to a leveraged set of transactions. This leveraging strategy, however, makes the set of transactions of the firm tightly coupled. And when any one transaction in a tightly coupled set fails, the set often collapses. Hence, a small business firm in this situation fails when a transaction in which it is highly dependent or leveraged happens to fail. The very strategies that contributed to overcoming the problems of liabilities of newness and small size turn out to be the root causes of failure. The experiences of the courseware companies are generalized to a process model of failure of new small businesses in turbulent environments.

The Complexity of Reliable Concurrency Control

We define what it means for a schedule to be reliable, that is, correct in the face of possible transaction failures (assuming that aborting a transaction to restore correctness is not allowed). It turns out that the right definition is recursive, and surprisingly involved. We show that, in fact, testing a schedule for reliability is PSPACE-complete, and thus in some sense even harder than the ordinary NP-complete notions of correctness examined in the past. However, we also prove that all conflict serializable schedules are always reliable, and thus reliability should not be an extra complication for practical concurrency control systems. Finally, we examine two other notions of reliability, related to multiple versions and aborts.

The Information Dilemma

Information asymmetries persist because in certain situations the value of information cannot be appropriated by the one who has it. A game-theoretic model of information asymmetries is presented, showing that under certain conditions there exists a prisoner's dilemma. The model is used to clarify the failure of markets to transmit certain types of information, to explain policy failure in regulation without recourse to government failure, and to identify new types of transaction failure in hierarchy and politics.

Principles of error detection in storage structures of database systems

The physical integrity of a database can be violated not only by transaction failures, system failures, and media failures, but also by undetected bugs in the hardware, the operating system, and the database management system (DBMS) itself. This paper presents online error detection techniques which can be built into the DBMS. First, a hierarchical model of a DBMS is briefly described. Subsequently, a classification of errors is outlined as a basis for the design of error detection techniques. Finally, objects and methods of online error detection are discussed for the hardware and the operating system, the DBMS buffer manager, record manager, and access path manager. Checking and maintaining the necessary redundancies requires only a slight CPU time overhead and no additional I/O operations.

Performance Modeling of Database Recovery Protocols

The performance modeling described in this paper compares several protocols which ensure that a database can be recovered to a consistent state after a transaction failure or system crash. The contributions of the paper include a collection of simple analytic models, based on Markov processes, for these protocols and some surprising results on the relative performance of the protocols. We consider only two-stage transactions (all reads before writes) and ignore effects of serializing transactions. The most interesting performance result presented is that, for systems obeying the assumptions of this paper, the "pessimistic" policy of holding write locks to commit point is considerably less efficient than the "optimistic" policy which allows reading of uncommitted data, but risks cascading aborts. A multiversion policy introduced in [2] was also studied and found always to be nearly as good as the optimistic policy and sometimes much better.

A database cache for high performance and fast restart in database systems

Performance in database systems is strongly influenced by buffer management and transaction recovery methods. This paper presents the principles of the database cache, which replaces the traditional buffer. In comparison to buffer management, cache management is more carefully coordinated with transaction management, and integrates transaction recovery. High throughput of small- and medium-sized transactions is achieved by fast commit processing and low database traffic. Very fast handling of transaction failures and short restart time after system failure are guaranteed in such an environment. Very long retrieval and update transactions are also supported.

Human Performance Engineering Considerations for Very Large Computer-Based Systems: The End User

Effective Human Performance Engineering for a large-scale, computer-based system involves many complex strategic and tactical decisions regarding the computer system design, the target user's behavior, and the organization/environment. Descriptions of the more important performance considerations are presented. These are based primarily on the experience accrued during the last several years in the building of Bell Laboratories centrally developed computer systems for use by telephone company loop operations personnel in assisting them to do their job. The target population can be broken down into these distinct classes: End Users, Database Maintainers, and the Data System Support Staff. This paper focuses on the End User, and specifically the Bell System Service Representative. Important points include: (i) early emphasis of human performance considerations in the computer system design process can reap valuable benefits; (ii) care must be taken to specify input/output design features which have gone through the human/system engineering step of identifying a favorable payoff versus penalty ratio; and (iii) based on measurement data and user interrogation, computer system availability and transaction failures and response times can seriously damage user performance and system acceptance.