Primary Copy Research Articles

QoS-Aware Fault-Tolerant Scheduling for Real-Time Tasks on Heterogeneous Clusters

Fault-tolerant scheduling plays a significant role in improving system reliability of clusters. Although extensive fault-tolerant scheduling algorithms have been proposed for real-time tasks in parallel and distributed systems, quality of service (QoS) requirements of tasks have not been taken into account. This paper presents a fault-tolerant scheduling algorithm called QAFT that can tolerate one node's permanent failures at one time instant for real-time tasks with QoS needs on heterogeneous clusters. In order to improve system flexibility, reliability, schedulability, and resource utilization, QAFT strives to either advance the start time of primary copies and delay the start time of backup copies in order to help backup copies adopt the passive execution scheme, or to decrease the simultaneous execution time of the primary and backup copies of a task as much as possible to improve resource utilization. QAFT is capable of adaptively adjusting the QoS levels of tasks and the execution schemes of backup copies to attain high system flexibility. Furthermore, we employ the overlapping technology of backup copies. The latest start time of backup copies and their constraints are analyzed and discussed. We conduct extensive experiments to compare our QAFT with two existing schemes-NOQAFT and DYFARS. Experimental results show that QAFT significantly improves the scheduling quality of NOQAFT and DYFARS.

Boosting adaptivity of fault-tolerant scheduling for real-time tasks with service requirements on clusters

Abstract Thank to the excellent extensibility and usability, computer clusters have become the dominating platform for parallel computing. Fault-tolerance is mandatory for safety-critical applications running on clusters. In this paper we propose a service-aware and adaptive fault-tolerant scheduling algorithm using overlapping technologies (SAO in short) that can tolerate a node’s permanent failure at any time instant for real-time tasks with service requirements in heterogeneous clusters. SAO adopts the primary/backup model and considers the timing constraints, service requirements, and system resource utilization. To improve system resource utilization, we employ backup- backup (BB in short) and primary- backup (PB in short) overlapping technologies and analyze the overlapping constraints. In addition, SAO has high system adaptivity by dynamically adjusting the service levels of tasks based on system load. Furthermore, to improve resource utilization and schedulability, SAO makes backup copies adopt passive execution scheme or decrease the overlapping execution time of the primary copy and backup copy of a task as much as possible. Compared with a baseline algorithm SAWO (a service-aware and adaptive fault-tolerant scheduling algorithm without using overlapping technologies) and an existing algorithm DYFARS with simulation experiments, SAO achieves an average of 51.25% improvement in performability.

Abstract 2155: Genetics of Primary and Metastatic Ovarian Cancer

Abstract Objective: The purpose of this study is to determine the differential copy number abnormalities (CNA) in primary serous epithelial ovarian cancer (EOC) and matched EOC metastatic specimens utilizing comparative genomic hybridization (CGH) technology. We are testing the hypothesis that comparing amplified and deleted regions that are similar or different between primary and metastatic may identify key drivers of metastasis and highlight genetic changes that occur during metastasis. Amplified regions lost in the metastatic tumor may indicate loci not required for progression and more aggressive disease. Finally, expression of genes in amplified and deleted regions will be measured to determine if their expression is indeed driven by the copy number. Materials: Primary and metastatic serous EOC tumors were collected from patients at the time of their initial cytoreductive surgery. Eligible patients were postmenopausal and presented with a pelvic mass, an elevated tumor marker (CA125 or HE4), signs of metastatic EOC on exam or imaging, and had never received chemotherapy. All women provided signed informed consent as part of an IRB-approved protocol. Primary and metastatic EOC specimens were confirmed by a dedicated gynecologic pathologist to be serous histology and were analyzed with the Agilent 4×180K aCGH platform, using DNA from each patient's blood as a matched control. The magnitude of deletions and amplifications in CNAs measured over the whole genome were determined by a combination of Nexus and in-house software. Results: Among the eight patients eligible for analysis, the median age was 68 years (range 54-78); six (75%) were Caucasian; one was Hispanic (12.5%) and one (12.5%) was African-American. All patients had high grade serous EOC. The majority of patients (87.5%) had stage IIIc disease, and one woman had stage IIc disease. Primary and metastatic serous EOC tumor DNA demonstrated frequent amplifications and deletions when compared to matched controls. Preliminary analyses of EOC tumors demonstrated amplifications and deletions unique to metastatic specimens as compared to their matched primary EOC tumors. Regions affected included amplifications and deletions known to be associated with therapeutic responses. Comparison to publically available larger primary tumor copy number data is revealing recurrent specific regions that are preserved in the metastatic tumors while others are not suggesting importance for tumor progression. Conclusions: Genomic analysis of EOC may elucidate pathways of carcinogenesis and metastasis. Genes that are amplified in primary tumors but disappear in corresponding metastases may represent regions that are not critical for tumor growth and progression. Genes that are amplified in metastatic implants may represent important loci for metastatic potential of EOC, and could ultimately represent potential targets for therapies. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2155.

Experience building and operating the CMS Tier-1 computing centres

The CMS Collaboration relies on 7 globally distributed Tier-1 computing centres located at large universities and national laboratories for a second custodial copy of the CMS RAW data and primary copy of the simulated data, data serving capacity to Tier-2 centres for analysis, and the bulk of the reprocessing and event selection capacity in the experiment. The Tier-1 sites have a challenging role in CMS because they are expected to ingest and archive data from both CERN and regional Tier-2 centres, while they export data to a global mesh of Tier-2s at rates comparable to the raw export data rate from CERN. The combined capacity of the Tier-1 centres is more than twice the resources located at CERN and efficiently utilizing this large distributed resources represents a challenge. In this article we will discuss the experience building, operating, and utilizing the CMS Tier-1 computing centres. We will summarize the facility challenges at the Tier-1s including the stable operations of CMS services, the ability to scale to large numbers of processing requests and large volumes of data, and the ability to provide custodial storage and high performance data serving. We will also present the operations experience utilizing the distributed Tier-1 centres from a distance: transferring data, submitting data serving requests, and submitting batch processing requests.

Comparing Genetic Characteristics of Retrotransposon TOS17 During Different Tissue Culture Processes in the Rice Cultivars Nipponbare and Shishoubaimao

As a retrotransposon, TOS17 was a useful tool for rice genetic and functional genomic research. To ascertain the feasibility of constructing a TOS17 insertion mutation library in the rice cultivar Shishoubaimao, the genetic and expression characteristics of TOS17 were analyzed. We made solid and suspension tissue cultures and confirmed the copy numbers of TOS17 at different time points in both tissue culture processes by real-time quantitative PCR (RT-qPCR). Three primary copies of TOS17 were detected in naturally grown Shishoubaimao. TOS17 was activated by tissue culture, and the copy numbers of TOS17 increased along with a prolonged tissue culture time in both the Nipponbare and the Shishoubaimao cultivars. Therefore, Shishoubaimao is a potential candidate for constructing a TOS17 insertion mutant library. Compared with Nipponbare, TOS17 was more active in Shishoubaimao during tissue culture. Higher copy numbers of TOS17 were obtained with the suspension tissue culture process than with the solid tissue culture process over the same time courses. We concluded that 3–4 months of suspension tissue culture time is suitable for constructing a TOS17 insertion mutant library in Shishoubaimao.

Improving Data Availability for Better Access Performance: A Study on Caching Scientific Data on Distributed Desktop Workstations

Client-side data caching serves as an excellent mechanism to store and analyze the rapidly growing scientific data, motivating distributed, client-side caches built from unreliable desktop storage contributions to store and access large scientific data. They offer several desirable properties, such as performance impedance matching, improved space utilization, and high parallel I/O bandwidth. In this context, we are faced with two key challenges: (1) the finite amount of contributed cache space is stretched by the ever increasing scientific dataset sizes and (2) the transient nature of volunteered storage nodes impacts data availability. In this article, we address these challenges by exploiting the existence of external, primary copies of datasets. We propose a novel combination of prefix caching, collective download, and remote partial data recovery (RPDR), to deal with optimal cache space consumption and storage node volatility. Our evaluation, performed on our FreeLoader prototype, indicates that prefix caching can significantly improve the cache hit rate and partial data recovery is better than (or comparable to) many persistent-data availability techniques.

Exploiting Redundancies to Enhance Schedulability in Fault-Tolerant and Real-Time Distributed Systems

In the past decades, distributed systems have been widely applied to real-time applications, most of which have fault-tolerance requirements to assure high reliability. Due to the stringent space constraints of real-time systems, the issue of schedulability becomes a major concern in the design of fault-tolerant and real-time distributed systems. Most existing real-time and fault-tolerant scheduling algorithms, which are based on the primary-backup scheme for periodic real-time tasks, introduce unnecessary redundancies by aggressively using active-backup copies. To solve this problem, we propose two novel fault-tolerant techniques, which are seamlessly integrated with fixed-priority-based scheduling algorithms. These techniques leverage redundancies to enhance schedulability in fault-tolerant and real-time distributed systems. Our fault-tolerant techniques make use of the primary-backup scheme to tolerate permanent hardware failures. The first technique (referred to as Tercos) terminates the execution of active-backup copies, when corresponding primary copies are successfully completed. Tercos is designed to reduce scheduling lengths in fault-free scenarios to enhance schedulability by virtue of executing portions of active-backup copies in passive forms. The second technique (referred to as Debus) uses a deferred-active-backup scheme to further minimize schedule lengths to improve the schedulability performance. Debus schedules active-backup copies as late as possible, while terminating active-backup copies when their primary copies are completed. Experimental results show that, compared with existing algorithms in literature, Tercos can significantly improve schedulability by up to 17.0% (with an average of 9.7%). Furthermore, empirical results reveal that Debus can enhance schedulability over Tercos by up to 12% (with an average of 7.8%).

On the Design of Fault-Tolerant Scheduling Strategies Using Primary-Backup Approach for Computational Grids with Low Replication Costs

Fault-tolerant scheduling is an imperative step for large-scale computational grid systems, as often geographically distributed nodes co-operate to execute a task. By and large, primary-backup approach is a common methodology used for fault tolerance wherein each task has a primary copy and a backup copy on two different processors. In this paper, we identify two cases that may happen when scheduling dependent tasks with primary-backup approach. We derive two important constraints that must be satisfied. Further, we show that these two constraints play a crucial role in limiting the schedulability and overloading efficiency of backups of dependent tasks. We then propose two strategies to improve schedulability and overloading efficiency, respectively. We propose two algorithms (MRC-ECT and MCT-LRC), to schedule backups of independent jobs and dependent jobs, respectively. MRC-ECT is shown to guarantee an optimal backup schedule in terms of replication cost for an independent task, while MCT-LRC can schedule a backup of a dependent task with minimum completion time and less replication cost. We conduct extensive simulation experiments to quantify the performance of the proposed algorithms.

SMAD proteins control DROSHA-mediated microRNA maturation

MicroRNAs (miRNAs) are small non-coding RNAs that participate in the spatiotemporal regulation of messenger RNA and protein synthesis. Aberrant miRNA expression leads to developmental abnormalities and diseases, such as cardiovascular disorders and cancer; however, the stimuli and processes regulating miRNA biogenesis are largely unknown. The transforming growth factor beta (TGF-beta) and bone morphogenetic protein (BMP) family of growth factors orchestrates fundamental biological processes in development and in the homeostasis of adult tissues, including the vasculature. Here we show that induction of a contractile phenotype in human vascular smooth muscle cells by TGF-beta and BMPs is mediated by miR-21. miR-21 downregulates PDCD4 (programmed cell death 4), which in turn acts as a negative regulator of smooth muscle contractile genes. Surprisingly, TGF-beta and BMP signalling promotes a rapid increase in expression of mature miR-21 through a post-transcriptional step, promoting the processing of primary transcripts of miR-21 (pri-miR-21) into precursor miR-21 (pre-miR-21) by the DROSHA (also known as RNASEN) complex. TGF-beta- and BMP-specific SMAD signal transducers are recruited to pri-miR-21 in a complex with the RNA helicase p68 (also known as DDX5), a component of the DROSHA microprocessor complex. The shared cofactor SMAD4 is not required for this process. Thus, regulation of miRNA biogenesis by ligand-specific SMAD proteins is critical for control of the vascular smooth muscle cell phenotype and potentially for SMAD4-independent responses mediated by the TGF-beta and BMP signalling pathways.

Performance modelling of asynchronous replica distribution technique on distributed database systems

In distributed database systems, replication is the most widely used approach to offering high data availability, increased fault tolerance, and improved scalability of the overall system. However, the primary concern with this technique is not only to provide high availability and reliability but also to boost the performance of data replication, i.e. to minimize response time and maximize throughput. We propose an analytical model for asynchronous replication technique (ART) by copying the data and replicating it to remote sites on a continuous basis. A remote copy of the data is created at a subsequent time after the primary copy is updated. The model is based on the M/G/1 queuing system by taking into account intersite communication between sites in the systems.

A novel fault-tolerant scheduling algorithm for precedence constrained tasks in real-time heterogeneous systems

Fault-tolerance is an essential requirement for real-time systems, due to potentially catastrophic consequences of faults. In this paper, we investigate an efficient off-line scheduling algorithm generating schedules in which real-time tasks with precedence constraints can tolerate one processor’s permanent failure in a heterogeneous system with fully connected network. The tasks are assumed to be non-preemptable, and each task has two copies scheduled on different processors and mutually excluded in time. In the literature in recent years, the quality of a schedule has been previously improved by allowing a backup copy to overlap with other backup copies on the same processor. However, this approach assumes that tasks are independent of one other. To meet the needs of real-time systems where tasks have precedence constraints, a new overlapping scheme is proposed. We show that, given two tasks, the necessary conditions for their backup copies to safely overlap in time with each other are (1) their corresponding primary copies are scheduled on two different processors, (2) they are independent tasks, and (3) the execution of their backup copies implies the failures of the processors on which their primary copies are scheduled. For tasks with precedence constraints, the new overlapping scheme allows the backup copy of a task to overlap with its successors’ primary copies, thereby further reducing schedule length. Based on a proposed reliability model, tasks are judiciously allocated to processors so as to maximize the reliability of heterogeneous systems. Additionally, times for detecting and handling of a permanent fault are incorporated into the scheduling scheme. We have performed experiments using synthetic workloads as well as a real world application. Simulation results show that compared with existing scheduling algorithms in the literature, our scheduling algorithm improves reliability by up to 22.4% (with an average of 16.4%) and achieves an improvement in performability, a measure that combines reliability and schedulability, by up to 421.9% (with an average of 49.3%).

Replication and retrieval strategies for resource-effective admission control in multi-resolution video servers

Video-on-demand (VOD) service requires balanced use of system resources, such as disk bandwidth and buffer, to accommodate more clients. The data retrieval size and data rates of video streams directly affect the utilization of these resources. Given the data rates which vary widely in multi-resolution video servers, we need to determine the appropriate data retrieval size to balance the buffer with the disk bandwidth. Otherwise, the server may be unable to admit new clients even though one of the resources is available for use. To address this problem, we propose the following new schemes that work together: (1) A replication scheme called Splitting Striping units by Replication (SSR). To increase the number of admitted clients, SSR defines two sizes of striping unit, which allow data to be stored on the primary and backup copies in different ways. (2) A retrieval scheduling method which combines the merits of existing SCAN and grouped sweeping scheme (GSS) algorithms to balance the buffer and disk bandwidth usage. (3) Admission control algorithms which decide whether to read data from the primary or the backup copy. The effectiveness of the proposed schemes is demonstrated through simulations. Results show that our schemes are able to cope with various workloads efficiently and thus enable the server to admit a much larger number of clients.

주사본 권한을 이용한 동적 트랜잭션 분배 알고리즘

데이터베이스 공유 시스템(Database Sharing System: DSS)은 고성능 트랜잭션 처리를 위해 제안된 시스템이다. DSS에서 고속의 통신망으로 연결된 노드들은 별도의 메모리와 운영체제를 가지며, 데이터베이스를 저장하고 있는 디스크는 모든 노드에 의해 공유된다. 그리고 빈번한 디스크 액세스를 피하기 위해 각 노드는 자신의 메모리 버퍼에 최근에 액세스한 페이지들을 캐싱한다. 본 논문에서는 DSS를 구성하고 있는 각 노드의 부하를 효과적으로 분산한 수 있는 동적 트랜잭션 분배 알고리즘을 제안한다. 제안한 알고리즘은 각 노드에 할당된 주사본 권한을 이용함으로써 노드별 참조 지역성을 지원하고, 그 결과 캐쉬 이용률을 증가하여 디스크 액세스 수를 최소화한다. 뿐만 아니라, 노드의 현재 부하를 고려하여 트랜잭션 분배 정책을 결정함으로써 특정 노드에 트랜잭션이 집중되는 것을 피한다. 제안된 알고리즘의 성능평가를 위해 시뮬레이션 실험을 수행하였으며, 실험결과 제안된 알고리즘이 기존 알고리즘들보다 트랜잭션 처리율에서 높은 성능을 보였다. 특히 트랜잭션 부하량이 높은 경우와 편중된 데이터 참조를 보이는 경우에 좋은 성능을 보였다. Database sharing system (DSS) refers to a system for high performance transaction processing. In DSS, the processing nodes are locally coupled via a high speed network and share a common database at the disk level. Each node has a local memory and a separate copy of operating system. To reduce the number of disk accesses, the node caches database pages in its local memory buffer. In this paper, we propose a dynamic transaction routing algorithm to balance the load of each node in the DSS. The proposed algorithm is novel in the sense that it can support node-specific locality of reference by utilizing the primary copy authority assigned to each node; hence, it can achieve better cache hit ratios and thus fewer disk I/Os. Furthermore, the proposed algorithm avoids a specific node being overloaded by considering the current workload of each node. To evaluate the performance of the proposed algorithm, we develop a simulation model of the DSS, and then analyze the simulation results. The results show that the proposed algorithm outperforms the existing algorithms in the transaction processing rate. Especially the proposed algorithm shows better performance when the number of concurrently executed transactions is high and the data page access patterns of the transactions are not equally distributed.

HLR Replication Protocols for Global Mobility Management in PCS Networks

The new trend for PCS networks is to provide mobile users with large-scale mobile capability across many service areas. In this scenario, global database management for PCS networks has become an increasingly important research issue. In this paper, we examined two replicated database strategies, single-replica (SR) and multiple-replica (MR), for large-scale mobility of per-user data management in personal communication networks. The SR strategy uses a single replica approach of HLR. The MR strategy replicates the per-user data of HLR in many regions. The two strategies are based a partial replication scheme, and a primary copy method is used to maintain replicas' consistency. Our numerical results show that the MR strategy outperforms the SR strategy in most situations; however, it may be worse when the probability of a mobile user visiting a foreign region is high and the query rates from other foreign regions are low. Additionally, the number of replicas should be compact in the MR strategy in order to achieve a reasonable query response time. Therefore, we propose an adaptive multiple replication protocol to choose a suitable replication strategy and to decide the optimized number of replicas.

Replica Consistency in Lazy Master Replicated Databases

In a lazy master replicated database, a transaction can commit after updating one replica copy (primary copy) at some master node. After the transaction commits, the updates are propagated towards the other replicas (secondary copies), which are updated in separate refresh transactions. A central problem is the design of algorithms that maintain replica's consistency while at the same time minimizing the performance degradation due to the synchronization of refresh transactions. In this paper, we propose a simple and general refreshment algorithm that solves this problem and we prove its correctness. The principle of the algorithm is to let refresh transactions wait for a certain “deliver time” before being executed at a node having secondary copies. We then present two main optimizations to this algorithm. One is based on specific properties of the topology of replica distribution across nodes. In particular, we characterize the nodes for which the deliver time can be null. The other improves the refreshment algorithm by using an immediate update propagation strategy.

Update propagation strategies to improve freshness in lazy master replicated databases

Many distributed database applications need to replicate data to improve data availability and query response time. The two-phase commit protocol guarantees mutual consistency of replicated data but does not provide good performance. Lazy replication has been used as an alternative solution in several types of applications such as on-line financial transactions and telecommunication systems. In this case, mutual consistency is relaxed and the concept of freshness is used to measure the deviation between replica copies. In this paper, we propose two update propagation strategies that improve freshness. Both of them use immediate propagation: updates to a primary copy are propagated towards a slave node as soon as they are detected at the master node without waiting for the commitment of the update transaction. Our performance study shows that our strategies can improve data freshness by up to five times compared with the deferred approach.

THE ADR REPLICATION MANAGER

ADR (Atomic Delayed Replication) is a controllable replication manager implemented on top of commercial distributed relational databases. ADR's goal is to enable various well-defined trade-offs between database coherence, throughput and response time in large database networks, e.g. for telecom applications. By combining a strategy for distributed database design with a specific replication protocol, ADR preserves the ACID properties with a controlled relaxation of coherence between primary and secondary copies. We first discuss formal characteristics of ADR, and present the implementation techniques required to realize these formal characteristics on top of commercial distributed database technology. Then, after reviewing a validated analytical performance model for the approach, we demonstrate its flexibility by summarizing experiences with two industrial ADR applications in telecommunications management, both jointly developed with Philips Laboratories. One is database support for the integrated operation and evolution of Intelligent Network telephone services, where secondary copies are held within a distributed database system optimized for throughput and availability during schema evolution. The other concerns database support for mobile phones in a City-wide DECT setting (Digital Enhanced Cordless Telecommunications), where secondary copies are held in main memory caches outside the DBMS.

Fault-tolerant rate-monotonic first-fit scheduling in hard-real-time systems

Hard-real-time systems require predictable performance despite the occurrence of failures. In this paper, fault tolerance is implemented by using a novel duplication technique where each task scheduled on a processor has either an active backup copy or a passive backup copy scheduled on a different processor. An active copy is always executed, while a passive copy is executed only in the case of a failure. First, the paper considers the ability of the widely-used rate-monotonic scheduling algorithm to meet the deadlines of periodic tasks in the presence of a processor failure. In particular, the completion time test is extended so as to check the schedulability on a single processor of a task set including backup copies. Then, the paper extends the well-known rate-monotonic first-fit assignment algorithm, where all the task copies, included the backup copies, are considered by rate-monotonic priority order and assigned to the first processor in which they fit. The proposed algorithm determines which tasks must use the active duplication and which can use the passive duplication. Passive duplication is preferred whenever possible, so as to overbook each processor with many passive copies whose primary copies are assigned to different processors. Moreover, the space allocated to active copies is reclaimed as soon as a failure is detected. Passive copy overbooking and active copy deallocation allow many passive copies to be scheduled sharing the same time intervals on the same processor, thus reducing the total number of processors needed. Simulation studies reveal a remarkable saving of processors with respect to those needed by the usual active duplication approach in which the schedule of the non-fault-tolerant case is duplicated on two sets of processors.

Multiview access protocols for large-scale replication

The article proposes a scalable protocol for replication management in large-scale replicated systems. The protocol organizes sites and data replicas into a tree-structured, hierarchical cluster architecture. The basic idea of the protocol is to accomplish the complex task of updating replicated data with a very large number of replicas by a set of related but independently committed transactions. Each transaction is responsible for updating replicas in exactly one cluster and invoking additional transactions for member clusters. Primary copies (one from each cluster) are updated by a cross-cluster transaction. Then each cluster is independently updated by a separate transaction. This decoupled update propagation process results in possible multiple views of replicated data in a cluster. Compared to other replicated data management protocols, the proposed protocol has several unique advantages. First, thanks to a smaller number of replicas each transaction needs to atomically update in a cluster, the protocol significantly reduces the transaction abort rate, which tends to soar in large transactional systems. Second, the protocol improves user-level transaction response time as top-level update transactions are allowed to commit before all replicas have been updated. Third, read-only queries have the flexibility to see database views of different degrees of consistency and data currency. This ranges from global, most up to date, and consistent views, to local, consistent, but potentially old views, to local, nearest to users but potentially inconsistent views. Fourth, the protocol maintains its scalability by allowing dynamic system reconfiguration as it grows by splitting a cluster into two or more smaller ones. Fifth, autonomy of the clusters is preserved as no specific protocol is required to update replicas within the same cluster. Clusters are, therefore, free to use any valid replication or concurrency control protocols.

A simulation model for the primary copy strategy in distributed INGRES

In distributed INGRES, the primary copy strategy based on two-phase locking is used to solve the concurrency control (cc) problem. To observe the behaviors of the transactions in distributed INGRES, a simulation model is proposed and a simulation is conducted in this paper. Furthermore, to observe how the primary copy sites for fragments influence the performance of distributed INGRES, we derive a method to select the primary copy sites such that the system has shorter average processing time per transaction, higher throughput, and fewer deadlocks. The experimental results obtained from the simulation show that the primary copy sites selected by the method generate the best results under low conflict ratio or long inter-arrival time.