As supercomputers increase in size, the mean time between failures (MTBF) of a system becomes shorter, and the reliability problem of supercomputers becomes more and more serious. MPI is currently the de facto standard used to build high-performance applications, and researches on the fault tolerance methods of MPI are always hot topics. However, due to the characteristics of MPI programs, most current checkpointing methods for MPI programs need to modify the MPI library (even operating system), or implement a complicated protocol by logging lots of messages. In this paper, we carry forward the idea of Application-Level Checkpointing (ALC). Based on the general fact that programmers are familiar with the communication characteristics of applications, we have developed BC-ALC, a new portable blocking coordinated ALC for MPI programs. BC-ALC neither modifies the MPI library (even operating system) nor logs any message. It implements coordination only by the Barrier operations instead of any complicated protocol. Furthermore, in order to reduce the cost of fault-tolerance, we reduce the synchronization range of the barrier, and design WBC-ALC, a weak blocking coordinated ALC utilizing group synchronization instead of global synchronization based on the communication relationship between processes. We also propose a fault-tolerance framework developed on top of WBC-ALC and discuss an implementation of it. Experimental results on NPB3.3-MPI benchmarks validate BC-ALC and WBC-ALC, and show that compared with BC-ALC, the average coordination time and the average backup time of a single checkpoint in WBC-ALC are reduced by 44.5% and 5.7% respectively.

MapReduce is commonly used as a parallel massive data processing model. When deploying it as a service over the open systems, the computational integrity of the participants is becoming an important issue due to the untrustworthy workers. Current duplication-based solutions can effectively solve non-collusive attacks, yet most of them require a centralized worker to re-compute additional sampled tasks to defend collusive attacks, which makes the worker a bottleneck. In this paper, we try to explore a trusted worker scheduling framework, named VAWS, to detect collusive attackers and assure the integrity of data processing without extra re-computation. Based on the historical results of verification, we construct an Integrity Attestation Graph (IAG) in VAWS to identify malicious mappers and remove them from the framework. To further improve the efficiency of identification, a verification-couple selection method with the IAG guidance is introduced to detect the potential accomplices of the confirmed malicious worker. We have proven the effectiveness of our proposed method on the improvement of system performance in theoretical analysis. Intensive experiments show the accuracy of VAWS is over 97% and the overhead of computation is closed to the ideal value of 2 with the increasing of the number of map tasks in our scheme.

Scaling up the system size has been the common approach to achieving high performance in parallel computing. However, designing and implementing a large-scale parallel system can be very costly in terms of money and time. When building a target system, it is desirable to initially build a smaller version by using the processing nodes with the same architecture as those in the target system. This allows us to achieve efficient and scalable prediction by using the smaller system to predict the performance of the target system. Such scalability prediction is critical because it enables system designers to evaluate different design alternatives so that a certain performance goal can be successfully achieved. As the de facto standard for writing parallel applications, MPI is widely used in large-scale parallel computing. By categorizing the discrete event simulation methods for MPI programs and analyzing the characteristics of scalability prediction, we propose a novel simulation method, called virtual-actual combined execution-driven (VACED) simulation, to achieve scalable prediction for MPI programs. The basic idea behind is to predict the execution time of an MPI program on a target machine by running it on a smaller system so that we can predict its communication time by virtual simulation and obtain its sequential computation time by actual execution. We introduce a model for the VACED simulation as well as the design and implementation of VACED-SIM, a lightweight simulator based on fine-grained activity and event definitions. We have validated our approach on a sub-system of Tianhe-1A. Our experimental results show that VACED-SIM exhibits higher accuracy and efficiency than MPI-SIM. In particular, for a target system with 1024 cores, the relative errors of VACED-SIM are less than 10% and the slowdowns are close to 1.

Computation integrity is difficult to verify when mass data processing is outsourced. Current integrity protection mechanisms and policies verify results generated by participating nodes within a computing environment of service providers (SP), which cannot prevent the subjective cheating of SPs. This paper provides an analysis and modeling of computation integrity for mass data processing services. A third-party sampling-result verification method, named TS-TRV, is proposed to prevent lazy cheating by SPs. TS-TRV is a general solution of verification on the intermediate results of common MapReduce jobs, and it utilizes the powerful computing capability of SPs to support verification computing, thus lessening the computing and transmission burdens of the verifier. Theoretical analysis indicates that TS-TRV is effective on detecting the incorrect results with no false positivity and almost no false negativity, while ensuring the authenticity of sampling. Intensive experiments show that the cheating detection rate of TS-TRV achieves over 99% with only a few samples needed, the computation overhead is mainly on the SP, while the network transmission overhead of TS-TRV is only O(log N).