Achieving high availability and performance computing with an HA-OSCAR cluster

The event service is the earliest CORBA solution to the message queue model of communication in distributed systems. Typical implementations, however, suffer from the lack of event delivery guarantees. The loss of messages is aggravated by the presence of burstiness in the input to the event service, and occurrences of isolated bursts of traffic could also have serious effects. In this paper, we develop stochastic reward net (SRN) models that can aid in the study and configuration of the event service to conform to design specifications. To capture burstiness in the input, a Markov-modulated Poisson process (MMPP) is used as the input source. Erlang distributed event consumption times are used in the models to accommodate more general distributions and a wider range of variances. The models also take into consideration the FIFO discard policy adopted in many event service implementations. The SRN models are solved using the tool SPNP (Stochastic Petri Net Package). The applicability of the models to the CORBA notification service is also briefly discussed.

Commercial off-the-shelf (COTS) software products are being included in ever more complex and critical systems. There are clearly advantages to considering the use of COTS in such systems, but given the rigorous needs of such critical systems or subsystems, there are concerns about the suitability of COTS software for such applications. The AIAA’s Software Systems Technical Committee and the AIAA’s Computer Systems Technical Committee have undertaken an effort to produce a guidebook that captures issues related to the consideration and use of COTS products in these large, complex systems, with a special emphasis on those risks and risk mitigation approaches that relate to mission critical systems. This guidebook identifies a set of characteristics of mission critical systems that makes the consideration, selection and validation processes of COTS products (hardware, software, subsystems, etc.) an emerging factor in systems definition, development and acceptance. The guidebook discusses a large number of risk areas related to using COTS software products in mission critical systems and identifies various mitigation approaches to the risks. It includes detailed processes for selecting and evaluating the products and considerations for using COTS software products within the overall context of the software development life cycle standardized in IEEE/EIA 12207. Mission critical system characteristics such as reliability, safety, availability, maintainability and certification tend to influence whether or not COTS software should be considered for a given application. Once the suitability of COTS software has been determined, then it is possible that additional requirements may be placed on the product and/or the product’s vendor for such mission critical applications. Further, it is possible that certain system requirements and expectations may need to be modified because of the inclusion of COTS software products in that system. As COTS products continue to be considered as candidates for inclusion within mission critical systems, there are likely to be additional concerns and factors to emerge that will influence how both acquirers and suppliers decide if and/or when to use COTS products. Ongoing monitoring of this technology area seems to be warranted.

How to predict the high availability (HA) using quantitative approach becomes important during system design phase. Stochastic reward nets (SRN) is used to model the behavior of the system. In this paper, we will adopt SRN as an analytic model to compute several interesting output measures that characterize availability features of a HA messaging system. This messaging service is the core part of a 24-hour non-stop safety-critical financial service. During the system design phase, this prediction model is used to provide useful information for aiding special technology choosing.

The reputation of network providers strongly depends on their ability to guarantee high performance levels of virtualized infrastructures, and to maintain strict Quality-of-Service (QoS) requirements, thus, the concept of nines or high availability (HA) is critical. It means that, on average, the continuity of a provided service cannot be violated for more than about five minutes per year. In this direction, we propose a framework useful to design and model, from a HA perspective, the Service Function Chains (SFCs) whose chained software logic is embodied in many softwarized telco infrastructures (e.g. IP Multimedia Subsystem elected here as a representative use case). The proposed framework interacts with TimeNET tool, and offers interesting functionalities such as: i) generating stochastic models of SFCs based on the SRN (Stochastic Reward Nets) formalism; ii) deploying network scenarios via drag-and-drop operations for basic users, or modifying the underlying SRN models for advanced users; iii) setting a variety of parameters (mean-time-to-failure/repair, software/hardware specs, redundancy, etc.); iv) presenting availability results in tabular and/or graphical forms.

Cluster computing has recently gone through an evolution from single processor systems to multicore/multi-socket systems. This has resulted in lowering the cost/performance ratio of the compute machines. Compute farms that host these machines tend to become heterogeneous over time due to incremental extensions, hardware upgrades and/or nodes being purchased for users with particular needs. This heterogeneity is not surprising given the wide range of processor, memory and network technologies that become available and the relatively small price difference between these various options. Different CPU architectures, memory capacities, communication and I/O interfaces of the participating compute nodes present many challenges to job scheduling and often result in under or over utilization of the compute resources. In general, it is not feasible for the application programmers to specifically optimize their programs for such a set of differing compute nodes, due to the difficulty and time-intensiveness of such a task. The trend of heterogeneous compute farms has coincided with resurgence in the virtualization technology. Virtualization technology is receiving widespread adoption, mainly due to the benefits of server consolidation and isolation, load balancing, security and fault tolerance. Virtualization has also generated considerable interest in the High Performance Computing (HPC) community, due to the resulting high availability, fault tolerance, cluster partitioning and accommodation of conflicting user requirements. However, the HPC community is still wary of the potential overheads associated with virtualization, as it results in slower network communications and disk I/O, which need to be addressed. The live migration feature, available to most virtualization technologies, can be leveraged to improve the throughput of a heterogeneous compute farm (HC) used for HPC applications. For this we mitigated the slow network communication in Xen; an open source virtual machine monitor. We present a detailed analysis of the communication framework of Xen and propose communication configurations that give 50% improvement over the conventional Xen network configuration. From a detailed study of the migration facility in Xen, we propose an improvement in the live migration facility specifically targeting HPC applications. This optimization gives around 50% improvement over the default migration facility of Xen. In this thesis, we also investigate resource scheduling in heterogeneous compute farm with the perspective of dynamic resource re-mapping. Our approach is to profile each job in the compute farm at runtime, and propose a better resource mapping compared to the initial allocation. We then migrate the job(s) to the best-suited homogeneous sub-cluster to improve overall throughput of the HC. For this, we develop a novel heterogeneity and virtualization-aware profiling framework, which is able to predict the CPU and communication characteristics of high performance scientific applications. Our experimental setup consists of a 32 node virtualized cluster with machines having CPU frequency between 2 GHz to 3 GHz. The framework was tested with a subset of NAS parallel benchmarks with the workload generated from Dror Feitelson’s workload archives. The prediction accuracy of our performance estimation model is over 80%. The framework implementation is lightweight, with an overhead of 3%. Our experiments show that we are able to improve the throughput of the compute farm by 25% and the time saved by the HC with our framework is over 30%. The framework can be readily extended to HCs supporting a cloud computing environment.

Stochastic Petri Net Package (SPNP) is a software package whose goal is to compute performance, availability or performability measures from Stochastic Petri Nets (SPN) and Fluid Stochastic Petri nets (FSPN). This software can use either analytic numeric methods, or simulation methods. Unfortunately, the standard discrete event simulation is inefficient to estimate the probabilities of rare events. For such rare event simulations, importance splitting technique is a good method to speedup the simulation. In the literature, two diffierent importance splitting techniques are known: RESTART and splitting. In this paper, we describe the application of these methods to (both fluid and discrete) Petri nets, their implementation in SPNP and we give some illustrations of the speed-up. The RESTART technique has already been applied in another Petri net package, TimeNet, but here we implement both RESTART and splitting, and we apply them to a more general class of Petri nets including the fluid ones.

Stream processing is widely used by today's applications such as financial data analysis and disaster response. In distributed stream processing systems, machine fail-stop events are handled by either active standby or passive standby. However, existing high availability (HA) schemes have not sufficiently addressed the situation when a machine becomes temporarily unavailable due to data rate spikes, intensive analysis or job sharing, which happens frequently but lasts for short time. It is not clear how well active and passive standby fare against such transient unavailability. In this paper, we first critically examine the suitability of active and passive standby against transient unavailability in a real testbed environment. We find that both approaches have advantages and drawbacks, but neither is ideal to provide fast recovery at low overhead as required to handle transient unavailability. Based on the insights gained, we propose a novel hybrid HA method that switches between active and passive standby modes depending on the occurrence of failure events. It presents a desirable tradeoff that is different from existing HA approaches: low overhead during normal conditions and fast recovery upon transient or permanent failure events. We have implemented our hybrid method and compared it with existing HA designs with comprehensive evaluation. The results show that our hybrid method can reduce two-thirds of the recovery time compared to passive standby and 80% message overhead compared to active standby, allowing applications to enjoy uninterrupted processing without paying a high premium.

Among the subsystems of railway, signalling system is the most important and key equipment of the automation system. Railway signalling system is responsible for directing and used to ensure the safety of train operation, to improve the efficiency of transport and other important tasks, thus the system has an irreplaceable role and high requirements of reliability. The graphical modeling ability of Stochastic Reward Net (SRN) for system reliability analysis can make the model intuitively described, at the same time, SRN can automatically generate a continuous time Markov chain, which has strong ability in mathematics. In this paper, firstly, the function and equipment structure of the railway signalling system is carried out, and then the reliability model of the system is established based on SRN to provide a reliable basis for the design and analysis of railway signalling system.

The stochastic Petri net package (SPNP) is a versatile modeling tool for the solution of generalized stochastic Petri nets (GSPN). The GSPN models are described in the input language for SPNP called C-based SPN language (CSPL). The CSPL is an extension of the C programming language with additional constructs that facilitate easy description of GSPN models. The GSPN models specified to SPNP are actually “GSPN reward models” that are based on the “Markov reward models” paradigm. This provides a powerful modeling environment for dependability analysis, performance analysis, and performability modeling. A number of important Petri net constructs like marking dependency, variable cardinality arc, and enabling functions facilitate the construction of models for complex systems.

An architecture of high performance computer,called Hyper Parallel Processing(HPP),is proposed to satisfy the requirements of both High Performance Computing,and Utility Computing which will be the application model of data centers.HPP combines the benefits of the scalability of MPP,the communication efficiency of DSM,as well as the commodity of cluster.Comparisons of current main-stream high performance computer architectures show that none of them can satisfy both HPC and utility computing.The main features of HPP are Global Address Space(GAS) and Hyper Node with single Operating System image.HPP supports the distributed global address space including both memory and I/O,but without hardware cache coherence.A Hyper Node consists of a set of application specific CPUs and a(or more) OS specific CPU.The OS CPU maintains the single system image,while the application CPUs run only lightweight run-time software.Besides the GAS interconnect network for applications,a standard SAN connects all OS CPUs and I/O devices providing system management and storage service.HPP is able to provide many opportunities of innovative research in High Performance and Utility Computing areas,including communication,synchronization,programming model,node operating system,utility computing,fault isolation,CPU and system etc.According to HPP architecture,a prototype system of Dawning5000 HPC is implemented and the feasibility of HPP is proved.

This Research to Practice Full Paper presents the experience of implementing micro HPC (μHPC) clusters in the education context. In developing countries, access to and experience of High Performance Computing (HPC) facilities (which are important for the economic development of those countries) has presented challenges to academic institutions due to prohibitive costs of acquisition and maintenance of HPC systems. These challenges affect the delivery of courses, which include HPC and parallel computation in their syllabi. However, due to advancements in integrated circuit technologies, credit-card-sized personal computers (PCs) have emerged with the capabilities of a fully-fledged PC. These credit-card-sized PCs offer an opportunity for system integrators to build micro HPC (μHPC) clusters which, apart from their efficiency, scalability and availability, have a number of attributes which are not shared with conventional data-center based HPC clusters These include their low power consumption, portability, diskless nodes, cost-effectiveness and excellent price to performance ratios. μHPC clusters are, therefore, ideally suited to support curricula in all educational disciplines that demand parallel computation. The purpose of this study is to investigate the implementation of μHPC to find out from students how ease of use of a μHPC system is to acquire skills and knowledge of HPC. We used focus groups and surveys to gather data in the framework of the design science research paradigm. The implementation of μHPC clusters involved a demonstration of the system in the problem context that provided valuable lessons. The results indicate that μHPC is the ease of use HPC artifact for learning about management, integration, deployment, installation, downloading, and running of parallel programs.

In the UNIX/Linux environment the kernel can log every command process created by every user with process accounting. Thus process accounting logs have many potential uses, particularly the monitoring and forensic investigation of security events. Previous work successfully leveraged the use of process accounting logs to identify a difficult to detect and damaging intrusions within high performance computing (HPC) clusters, masquerade attacks, where intruders pose as legitimate users with purloined authentication credentials. This paper incrementally advances the goal of more accurately identifying masqueraders on HPC clusters by seeking to identify features within command sets that distinguish masqueraders. To accomplish this goal, we created NVision-PA, a software tool which produces text and graphic statistical summaries describing input processing accounting logs. This research is both a promising next step toward creating a real-time masquerade detection sensor for production HPC clusters as well as providing another tool for system administrators to use for statistically monitoring and managing legitimate workloads in HPC environments

HDFS (Hadoop Distributed File System) is the primary storage of Hadoop. Even though data locality offered by HDFS is important for Big Data applications, HDFS suffers from huge I/O bottlenecks due to the tri-replicated data blocks and cannot efficiently utilize the available storage devices in an HPC (High Performance Computing) cluster. Moreover, due to the limitation of local storage space, it is challenging to deploy HDFS in HPC environments. In this paper, we present a hybrid design (Triple-H) that can minimize the I/O bottlenecks in HDFS and ensure efficient utilization of the heterogeneous storage devices (e.g. RAM, SSD, and HDD) available on HPC clusters. We also propose effective data placement policies to speed up Triple-H. Our design integrated with parallel file system (e.g. Lustre) can lead to significant storage space savings and guarantee fault-tolerance. Performance evaluations show that Triple-H can improve the write and read throughputs of HDFS by up to 7x and 2x, respectively. The execution times of data generation benchmarks are reduced by up to 3x. Our design also improves the execution time of the Sort benchmark by up to 40% over default HDFS and 54% over Lustre. The alignment phase of the Cloudburst application is accelerated by 19%. Triple-H also benefits the performance of SequenceCount and Grep in PUMA over both default HDFS and Lustre.

With the increasing availability of inexpensive yet powerful personal computers and servers, there has been an increasing trend toward personal computer cluster systems for network services. In this paper, we present a small-scale personal computer based cluster system for network servers as a low-cost alternative to traditional high-performance computing systems. The enhanced cluster system for scalable network services (CSSNS) consists of the parallel virtual file system (PVFS), the Linux virtual server (LVS), the director, and several high-end Pentium PC connected by high-speed switched Ethernet networks for I/O nodes and cluster nodes. The design and development of the cluster system for scalable network services is characterized by its high performance, high availability, high scalable file system, low cost, load balance on cluster servers, and convenient managing/setup for a broad range of applications. The transmission performance experiments for the entire system further validates and confirms that CSSNS is truly a self-made high-performance, high reliability and scalability personal computer based cluster system for scalable network servers.

Analytical modeling of TCP flow in wireless LANs