Mixed Granularity Research Articles

FAMG: A flow-aware and mixed granularity method for load-balancing in data center networks

In current Data center Networks (DCNs), Equal-Cost Multipath (ECMP) is a default load-balancing scheme. However, using ECMP may result in the rapid growth of Flow Completion Time (FCT) due to its well-known drawbacks. In order to solve the problems of ECMP, load balancing schemes with per-packet granularity (such as RPS) and per-flowlet granularity (such as LetFlow) are proposed. These two different granularities have their features, but there is no existing load-balancing scheme considering both their advantages. Moreover, these schemes ignore the traffic characteristics in DCNs. In this paper, we mix per-flowlet and per-packet granularity in scheduling decisions and consider traffic characteristics. Based on this idea, we propose a flow-aware and mixed granularity method for load-balancing, referred to as FAMG. The key idea of FAMG is to distinguish elephant and mouse flows and schedule them with different granularities. We evaluate FAMG in simulations with NS-3. Experimental results show that FAMG slightly sacrifices elephant flows FCT but it is worth it since our results demonstrate that FAMG is a good trade-off that outperforms RPS, LetFlow, and TLB in terms of mouse flows FCT in both symmetric and asymmetric network topology.

Automatic Repair of Java Programs with Mixed Granularity and Variable Mapping

During the process of software repair, since the granularity of repair is too coarse and the way of fixing ingredient is too simple, the repair efficiency needs to be further improved. To resolve the problems, we propose a Mixed Granularity and Variable Mapping based automatic software Repair (MGVMRepair). We adopt random search algorithm as the framework of program evolution, and utilize the mapping relationship between variables as an auxiliary specification. Firstly, fault localization is used to locate the suspicious statements and to form a list of modification points. Secondly, the ingredient of program repair at statement level is obtained, and the mapping relationship of variables is established. Then, the test case prioritization is improved from the perspective of the modification point. Finally, a program passes all test cases or the program iteration terminates. The experimental results show that MGVMRepair has a higher repair success rate than GenProg, CapGen, SimFix, jKali, jMutRepair and SketchFix on Defects4J.

The Surface Morphology of Large Nonpedunculated Colonic Polyps Predicts Synchronous Large Lesions

Large (≥20 mm) nonpedunculated colorectal polyps (LNPCPs) may have synchronous LNPCPs in up to 18% of cases. The nature of this relationship has not been investigated. We aimed to examine the relationship between individual LNPCP characteristics and synchronous colonic LNPCPs. Consecutive patients referred for resection of LNPCPs over 130 months until March 2022 were enrolled. Serrated lesions and mixed granularity LNPCPs were excluded from analysis. Patients with multiple LNPCPs resected were identified, and the largest was labelled as dominant. The primary outcome was the identification of individual lesion characteristics associated with the presence of synchronous LNPCPs. There were 3149 of 3381 patients (93.1%) who had a single LNPCP. In 232 (6.9%) a synchronous lesion was detected. Solitary lesions had a median size of 35 mm with a predominant Paris 0-IIa morphology (42.9%) and right colon location (59.5%). In patients with ≥2 LNPCPs, the dominant lesion had a median size of 40 mm, Paris 0-IIa (47.6%) morphology, and right colon location (65.9%). In this group, 35.8% of dominant LNPCPs were non-granular compared with 18.7% in the solitary LNPCP cohort. Non-granular (NG)-LNPCPs were more likely to demonstrate synchronous disease, with left colon NG-LNPCPs demonstrating greater risk (odds ratio, 4.78; 95% confidence interval, 2.95-7.73) than right colon NG-LNPCPs (odds ratio, 1.99; 95% confidence interval, 1.39-2.86). We found that 6.9% of LNPCPs have synchronous disease, with NG-LNPCPs demonstrating a greater than 4-fold increased risk. With post-colonoscopy interval cancers exceeding 5%, endoscopists must be cognizant of an individual's LNPCP phenotype when examining the colon at both index procedure and surveillance. gov, NCT01368289; NCT02000141; NCT02198729.

Improving the grinding performance of high-speed rail grinding stone via mixed granularity of zirconia corundum

High-speed rail grinding stones (GS) were fabricated using zirconia alumina (ZA) with five mass ratios of F16 to F30 to explore the regulatory mechanism of abrasives with mixed granularity on the grinding performance. With the decrease of abrasive grit size, the enhanced abrasive crushing strength can control the grinding behaviors of GS. GS with low mass ratio of F16 to F 30 gave the largest grinding amount (4.0 g) and grinding ratio (46.2), and the largest amounts of scratches on the rail surface, evincing that the fiercest grinding actions were realized via the mixed granularity, the improved strength and grinding capacity. A grinding equilibrium state was achieved via attrition-wear/micro-fracture of F16 and attrition-wear/pulled-out of F30, thus offering the best self-sharpness of GS.

Using data-mining techniques to improve combinatorial optimization algorithms

In this work, we show how data-mining can be used to cluster algorithmic generated data and use that data to improve algorithms that solve combinatorial optimization problems for a real-world application—the field-programmable gate array placement problem. Our methodology is a means for other algorithm engineers to improve their own algorithms for specific real-world problems that are hard to improve. In our case, the placement algorithms are difficult to improve, and to find better heuristics we analyze the results of placement solutions to find clustered information which can then be used to improve the algorithms. Specifically, we show a technique for gathering cluster information about placement, we create a new simulated annealing algorithm and a new genetic algorithm that can deal with a mixed granularity of placement objects on a virtual field-programmable gate array, and we show that these algorithms either execute faster or improve the overall quality of solution compared to their basic algorithm without this clustering data and improved heuristics. For our improved simulated annealing placer we improve the algorithms run-time by 17% across a range of benchmarks, and our genetic algorithm improves placement metrics—-critical path by 10% and channel-width by 4%.

Research on the simulation of PF-LBM model based on MPI+CUDA mixed granularity parallel

A microstructure numerical model is an intensive computational problem, for which the simulation time is too long and the simulation scale is too small. To solve these two problems, in this article, we use MPI+CUDA hybrid particle heterogeneous parallel computing to implement the dendrite growth simulation of a PF-LBM phase-field 3D model. Message Passing Interface (MPI) can be used to conduct coarse granularity division, to break through the limitation of the simulate scale in a single machine. In each node, fine-grained division is implemented by the Compute Unified Device Architecture (CUDA) parallel way to realize the completely parallelism intra-node, and to improve overall computational efficiency. At the same time, in this article, the "pseudo three-dimensional array" programming method is brought up in CUDA programming, and also to improve the CUDA random number generation method, in order to simplify the CUDA array programming and reduce the CUDA random number generation time purposes. Experiments show that at the same simulation scale, the speed-up ratio with 21 nodes MPI+CUDA was 57, which was increased 54% over the 21 nodes MPI. Under the condition of computing efficiency close, the largest simulation scale with 21 nodes MPI+CUDA was 4203, which is 13 times to single GPU. Therefore, the MPI + CUDA hybrid granularity parallel method proposed in this paper also has the advantages of high computational efficiency of the GPU and MPI to expand the simulation scale.

Multi-domain traffic partition grooming in mixed granularity optical networks

In wavelength-division-multiplexing (WDM) optical networks, the demands characterized by coarse and fine granularity cause the poor resource utilization of lightpaths in traffic grooming and an excessive utilization of (de)-multiplexing ports in waveband switching. More specifically, once a lightpath carries a coarse-granularity demand with a near-wavelength capacity, it will not be able to hold other demands; once a waveband carries a large amount of fine-granularity demands each with a very small wavelength fraction, the number of (de)-multiplexing ports will increase sharply. We call these inefficient situations as gra-diversity problems. Furthermore, with the large scale of the backbone, the optical network has been divided into multiple domains for achieving the scalability in the real world. Thus in this paper, we study gra-diversity problems and facilitate the novel traffic partition grooming with the granularity-layered graph (GLG) in the mixed granularity and multi-domain optical networks. Our heuristic is composed of three basic components: (a) the traffic grooming of fine-grained demands, (b) the waveband switching of coarse-grained demands, and (c) the demand migration that is achieved through the interconnection of different granular layers in GLG. Extensive simulation results demonstrate the effectiveness of our method on improving the resource utilization and mitigating port consumptions by well solving gra-diversity problems in the mixed granularity and multi-domain optical networks. From the comparisons between our method and benchmarks, we find that the average improvement ratio of blocking rate arrives at 34%, and the average ratio of port savings is about 3%.

Investigation of age-differentiated spatial semantic elaboration strategies for communicating route instructions

This paper investigates the effect of spatial semantic elaboration strategies on young and older participants' performance and preferences in a route drawing task where participants read route instructions and then drew the route on a map. In particular, an elaborated form of spatial perspective that communicated the route instruction in both egocentric and allocentric perspectives was compared against just the egocentric or allocentric perspective instructions individually. Additionally, route instruction granularity was varied to compare elaborated hierarchical instructions that communicated goal and landmark information with flat instructions. The results of the experiment showed that older participants performed best with least confusion in the allocentric perspective with flat granularity, but that they benefited from semantic elaboration when less optimal spatial strategies were used, unlike young participants who performed best with the mixed (i.e. elaborated) perspective and flat granularity. The experiment showed that older participants actually suffered from what might be information overload when excess semantic elaboration was provided, as was likely the case with the use of the mixed perspective, which confused them considerably. In addition, hierarchical granularity was only beneficial in confusing spatial perspectives, while it detracted from older participants' performance in their optimal perspective, showing that there is a fine balance to be struck between beneficial semantic elaboration and information overload.

Prediction of Partition Coefficients by Multiscale Hybrid Atomic-Level/Coarse-Grain Simulations

Coarse-grain models are becoming an increasingly important tool in computer simulations of a wide variety of molecular processes. In many instances it is, however, desirable to describe key portions of a molecular system at the atomic level. There is therefore a strong interest in the development of simulation methodologies that allow representations of matter with mixed granularities in a multiscale fashion. We report here a strategy to conduct mixed atomic-level and coarse-grain simulations of molecular systems with a recently developed coarse-grain model. The methodology is validated by computing partition coefficients of small molecules described in atomic detail and solvated by water or octane, both of which are represented by coarse-grain models. Because the present coarse-grain force field retains electrostatic interactions, the simplified solvent particles can interact realistically with the all-atom solutes. The partition coefficients computed by this approach rival the accuracy of fully atomistic simulations and are obtained at a fraction of their computational cost. The present methodology is simple, robust and applicable to a wide variety of molecular systems.

Efficiently supporting temporal granularities

Granularity is an integral feature of temporal data. For instance, a person's age is commonly given to the granularity of years and the time of their next airline flight to the granularity of minutes. A granularity creates a discrete image, in terms of granules, of a (possibly continuous) time-line. We present a formal model for granularity in temporal operations that is integrated with temporal indeterminacy, or "don't know when" information. We also minimally extend the syntax and semantics of SQL-92 to support mixed granularities. This support rests on two operations, scale and cast, that move times between granularities, e.g., from days to months. We demonstrate that our solution is practical by showing how granularities can be specified in a modular fashion, and by outlining a time- and space-efficient implementation. The implementation uses several optimization strategies to mitigate the expense of accommodating multiple granularities.

Space/time-efficient scheduling and execution of parallel irregular computations

In this article we investigate the trade-off between time and space efficiency in scheduling and executing parallel irregular computations on distributed-memory machines. We employ acyclic task dependence graphs to model irregular parallelism with mixed granularity, and we use direct remote memory access to support fast communication. We propose new scheduling techniques and a run-time active memory management scheme to improve memory utilization while retaining good time efficiency, and we provide a theoretical analysis on correctness and performance. This work is implemented in the context of the RAPID system which uses an inspector/executor approach to parallelize irregular computations at run-ti me. We demostrate the effectiveness of the proposed techniques on several irregular applications such as sparse matrix code and the fast multipole method for particle simulation. Our experimental results on Cray-T3E show that problems large sizes can be solved under limited space capacity, and that the loss of execution efficiency caused by the extra memory management overhead is reasonable.

Space and time efficient execution of parallel irregular computations

Solving problems of large sizes is an important goal for parallel machines with multiple CPU and memory resources. In this paper, issues of efficient execution of overhead-sensitive parallel irregular computation under memory constraints are addressed. The irregular parallelism is modeled by task dependence graphs with mixed granularities. The trade-off in achieving both time and space efficiency is investigated. The main difficulty of designing efficient run-time system support is caused by the use of fast communication primitives available on modern parallel architectures. A run-time active memory management scheme and new scheduling techniques are proposed to improve memory utilization while retaining good time efficiency, and a theoretical analysis on correctness and performance is provided. This work is implemented in the context of RAPID system [5] which provides run-time support for parallelizing irregular code on distributed memory machines and the effectiveness of the proposed techniques is verified on sparse Cholesky and LU factorization with partial pivoting. The experimental results on Cray-T3D show that solvable problem sizes can be increased substantially under limited memory capacities and the loss of execution efficiency caused by the extra memory managing overhead is reasonable.

Applying object-oriented technologies in modeling and querying temporally oriented clinical databases dealing with temporal granularity and indeterminacy.

The need of managing temporal information given at different levels of granularity or with indeterminacy is common to many application areas. Among them, we focus on clinical data management. Different time granularities and indeterminacy are also needed in querying temporal databases. In this paper, we describe GCH-OSQL (Granular Clinical History-Object Structured Query Language), an object-oriented temporally-oriented extension of SQL. GCH-OSQL is based on an object-oriented temporal data model, GCH-OODM. GCH-OODM allows storage of clinical information at different and mixed granularities or with temporal indeterminacy. GCH-OSQL deals with the valid time of clinical information. The temporal extension of the SELECT construct includes the addition of the TIME-SLICE and MOVING WINDOW clauses, and the capability to reference the temporal dimension of objects in the WHERE and SELECT clauses. Using object-oriented technologies, a system prototype for GCH-OSQL and GCH-OODM has been implemented and applied to data management of follow-up patients after coronary angioplasty intervention.

Run-Time Techniques for Exploiting Irregular Task Parallelism on Distributed Memory Architectures

Automatic scheduling for directed acyclic graphs (DAG) and its applications for coarse-grained irregular problems such as largen-body simulation have been studied in the literature. However, solving irregular problems with mixed granularities such as sparse matrix factorization is challenging since it requires efficient run-time support to execute a DAG schedule. In this paper, we investigate run-time optimization techniques for executing general asynchronous DAG schedules on distributed memory machines and discuss an approach for exploiting parallelism from commuting operations in the DAG model. Our solution tightly integrates the run-time scheme with a fast communication mechanism to eliminate unnecessary overhead in message buffering and copying. We present a consistency model incorporating the above optimizations, and take advantage of task dependence properties to ensure the correctness of execution. We demonstrate the applications of this scheme in sparse matrix factorizations and triangular equation solving for which actual speedups are difficult to obtain. We provide a detailed experimental study on Meiko CS-2 to show that the automatically scheduled code has achieved good performance for these difficult problems, and the run-time overhead is small compared to total execution times.