摘要 :
Structured business processes are the veins of complex business organizations. Workflows have generally been accepted as a means to model and support these processes, be they interactive or completely automated. The fact that thes...
展开
Structured business processes are the veins of complex business organizations. Workflows have generally been accepted as a means to model and support these processes, be they interactive or completely automated. The fact that these processes require robustness and clear semantics has generally been observed and has led to the combination of workflow and transaction concepts. Many variations on this combination exist, leading to many approaches to transactional workflow support. No clear classification of these approaches has been developed, however, resulting in a badly understood field. To deal with this problem, we describe a clear taxonomy of transactional workflow models in this paper, based on the relation between workflow and transaction concepts. We show that the classes in the taxonomy can be directly related to specification language and architecture types for workflow and transaction management systems. We compare the various classes with respect to their characteristics and place existing approaches in the taxonomy. We cover both "traditional" workflow approaches and more recent web-based approaches, including inter-organizational workflow approaches. Together, this paper offers a well-structured and concise analysis of the field of transactional business process support.
收起
摘要 :
Transactions have been around since the Seventies to provide reliable information processing in automated information systems. Originally developed for simple 'debit-credit' style database operations in centralized systems, they h...
展开
Transactions have been around since the Seventies to provide reliable information processing in automated information systems. Originally developed for simple 'debit-credit' style database operations in centralized systems, they have moved into much more complex application domains including aspects like distribution, process-orientation and loose coupling. The amount of published research work on transactions is huge and a number of overview papers and books already exist. A concise historic analysis providing an overview of the various phases of development of transaction models and mechanisms in the context of growing complexity of application domains is still missing, however. To fill this gap, this paper presents a historic overview of transaction models organized in several 'transaction management eras', thereby investigating numerous transaction models ranging from the classical flat transactions, via advanced and workflow transactions to the Web Services and Grid transaction models. The key concepts and techniques with respect to transaction management are investigated. Placing well-known research efforts in historical perspective reveals specific trends and developments in the area of transac-tion management. As such, this paper provides a comprehensive, structured overview of developments in the area.
收起
摘要 :
Transaction schedulers reduce the number of transaction reexecutions in applications using Software Transactional Memory (STM) by preventing conflicting transactions to run in parallel. Unfortunately, current scheduling solutions ...
展开
Transaction schedulers reduce the number of transaction reexecutions in applications using Software Transactional Memory (STM) by preventing conflicting transactions to run in parallel. Unfortunately, current scheduling solutions are too conservative, rely on coarse measures to serialize transactions, and are specially inadequate for workloads with long transactions. In this paper we introduce an optimistic and adaptive transaction scheduler that takes advantage of the information already collected by the STM runtime to increase parallelism between transactions and, thus, improve transactions' throughput. Our new ProVIT scheduler implements a low-overhead scheduling policy for short transactions that reduces concurrency as contention increases and a fine-grained scheduling policy for long transactions based on the novel concept of Very Important Transaction. Experimental results conducted with the STMBench7 benchmark and the STAMP benchmark suite showed that the ProVIT scheduler has comparable performance to other current scheduling solutions for short transactions, but is up to 65% faster for long-running transactions.
收起
摘要 :
This paper introduces the Mobile Nested Transactions (MNT) model for transaction processing coordination in groups of mobile devices. The model extends traditional approaches-centralized and distributed-for distributed transaction...
展开
This paper introduces the Mobile Nested Transactions (MNT) model for transaction processing coordination in groups of mobile devices. The model extends traditional approaches-centralized and distributed-for distributed transaction processing over mobile devices. Under the Nested Transactions approach, sub-transactions-logical work units-processing control is highly flexible and fault-tolerant. Hence, transaction processing with mobile groups lacking of access to a wired network or the Internet becomes possible. A MNT additional advantage is that sub-transactions could be distributed among mobile devices. This is, whenever some sub-transactions are successfully committed, they cannot be affected by failed or pending sub-transactions.
收起
摘要 :
Purpose - The purpose of this paper is to develop a step-by-step procedure, referred to as transaction formalism protocol (TFP) that the transaction development personnel will use to formalise transactions (communications) in the ...
展开
Purpose - The purpose of this paper is to develop a step-by-step procedure, referred to as transaction formalism protocol (TFP) that the transaction development personnel will use to formalise transactions (communications) in the domain of infrastructure management. The protocol is developed at two levels of abstraction: TFP specification and TFP tool. This paper presents the TFP specification in detail and introduces the TFP tool briefly. The specific focus of this paper is on the development process of the protocol specification. Design/methodology/approach - A four-step approach is used to develop the TFP; including, identify and select existing standards, benchmark standards, link and build on these standards, and develop TFP. To develop the protocol, the function modelling standard, integration definition function modelling (IDEFO) is used. The IDEFO treats each step of the protocol as a function. Findings - The TFP specification and TFP tool are developed using the proposed methodology. The TFP specification specifies inputs, controls, mechanisms, tools/techniques, and outputs required in each step, whereas the TFP tool defines forms for each step of the protocol that the transaction development personnel will use to define transactions in the domain of infrastructure management. Practical implications - The development of the TFP would enable the transaction development personnel (including transaction analysts, transaction designers, software developers, process modellers, and industry experts) to formalise transactions effectively and efficiently for the development of ICT-based collaboration systems. Originality/value - The proposed protocol incorporates shortcomings of existing standards. In contrast to other design standards that focus on either design or design cum implementation of the work processes and communications, the proposed TFP includes transaction monitoring and improvements in addition to the design and implementation of communications. Unlike other standards, the TFP is a detailed step-by-step procedure to ease its usability and understandability.
收起
摘要 :
Transactional memory (TM) has been proposed to alleviate some key programmability problems in chip multiprocessors. Most TMs optimistically allow concurrent transactions, detecting read-write or write-write conflicts. Upon conflic...
展开
Transactional memory (TM) has been proposed to alleviate some key programmability problems in chip multiprocessors. Most TMs optimistically allow concurrent transactions, detecting read-write or write-write conflicts. Upon conflicts, existing hardware TMs (HTMs) use one of three conflict-resolution policies: (1) always-abort, (2) always-wait for some conflicting transactions to complete, or (3) always-go past conflicts and resolve acyclic conflicts at commit or abort upon cyclic dependencies. While each policy has advantages, the policies degrade performance under contention by limiting concurrency (always-abort, always-wait) or incurring late aborts due to cyclic dependencies (always-go). Thus, while always-go avoids acyclic aborts, no policy avoids cyclic aborts. We propose Wait-n-GoTM (WnGTM) to increase concurrency while avoiding cyclic aborts. We observe that most cyclic dependencies are caused by threads interleaving multiple accesses to a few heavily-read-write-shared delinquent data cache blocks. These accesses occur in code sections called cycle inducer sections (CISTs). Accordingly, we propose Wait-n-Go (WnG) conflict-resolution to avoid many cyclic aborts by predicting and serializing the CISTs. To support the WnG policy, we extend previous HTMs to (1) allow multiple readers and writers, (2) scalably identify dependencies, and (3) detect cyclic dependencies via new mechanisms, namely, conflict transactional state, order-capture, and hardware timestamps, respectively. In 16-core simulations of STAMP, WnGTM achieves average speedups of 46% for higher-contention benchmarks and 28% for all benchmarks over always-abort (TokenTM) with low-contention benchmarks remaining unchanged, compared to always-go (DATM) and always-wait (LogTM-SE), which perform worse than and 6% better than TokenTM, respectively.
收起
摘要 :
Transactional memory (TM) has been proposed to alleviate some key programmability problems in chip multiprocessors. Most TMs optimistically allow concurrent transactions, detecting read-write or write-write conflicts. Upon conflic...
展开
Transactional memory (TM) has been proposed to alleviate some key programmability problems in chip multiprocessors. Most TMs optimistically allow concurrent transactions, detecting read-write or write-write conflicts. Upon conflicts, existing hardware TMs (HTMs) use one of three conflict-resolution policies: (1) alwaysabort, (2) always-wait for some conflicting transactions to complete, or (3) always-go past conflicts and resolve acyclic conflicts at commit or abort upon cyclic dependencies. While each policy has advantages, the policies degrade performance under contention by limiting concurrency (always-abort, always-wait) or incurring late aborts due to cyclic dependencies (always-go). Thus, while always-go avoids acyclic aborts, no policy avoids cyclic aborts. We propose Wait-n-GoTM (WnGTM) to increase concurrency while avoiding cyclic aborts. We observe that most cyclic dependencies are caused by threads interleaving multiple accesses to a few heavily-read-write-shared delinquent data cache blocks. These accesses occur in code sections called cycle inducer sections (CISTs). Accordingly, we propose Wait-n-Go (WnG) conflict-resolution to avoid many cyclic aborts by predicting and serializing the CISTs. To support the WnG policy, we extend previous HTMs to (1) allow multiple readers and writers, (2) scalably identify dependencies, and (3) detect cyclic dependencies via new mechanisms, namely, conflict transactional state, order-capture, and hardware timestamps, respectively. In 16-core simulations of STAMP, WnGTM achieves average speedups of 46% for higher-contention benchmarks and 28% for all benchmarks over always-abort (TokenTM) with low-contention benchmarks remaining unchanged, compared to always-go (DATM) and always-wait (LogTM-SE), which perform worse than and 6% better than TokenTM, respectively.
收起
摘要 :
One way that software transactional memory implementations attempt to reduce synchronization conflicts among transactions is by supporting different kinds of access modes. One such implementation, Dynamic Software Transactional Me...
展开
One way that software transactional memory implementations attempt to reduce synchronization conflicts among transactions is by supporting different kinds of access modes. One such implementation, Dynamic Software Transactional Memory (DSTM), supports three kinds of memory access: WRITE access, which allows an object to be observed and modified, READ access, which allows an object to be observed but not modified, and RELEASE access, which allows an object to be observed for a limited duration. In this paper, we examine the relative performance of these modes for simple benchmarks on a small-scale multiprocessor. We find that on this platform and for these benchmarks, the READ and RELEASE access benchmarks do not substantially increase transaction throughput (and sometimes reduce it). We blame the extra bookkeeping inherent in these access modes. In response, we propose a new SNAP access mode. This mode provides almost the same behavior as RELEASE mode, but admits much more efficient implementations.
收起
摘要 :
Distributed Transactional Memory (DTM) is a recent but promising model for programming distributed systems. It aims to present programmers with a simple to use distributed concurrency control abstraction (transactions), while main...
展开
Distributed Transactional Memory (DTM) is a recent but promising model for programming distributed systems. It aims to present programmers with a simple to use distributed concurrency control abstraction (transactions), while maintaining performance and scalability similar to distributed fine-grained locks. Any complications usually associated with such locks (e.g., distributed deadlocks) are avoided. In this article, we analyze the use of open nesting in the DTM setting. We extend two DTM algorithms, Transactional Forwarding Algorithm (TFA) and SCORe with support for open nested transactions and we implement them into two frameworks for running distributed transactions, such as Hyflow and Infinispan. We discuss the mechanisms and performance implications of such nesting, and identify the cases where using open nesting is warranted and the relevant parameters for such a decision. To the best of our knowledge, our work also contributes the first ever implementations of DTM systems with support for open-nested transactions.
收起