Objective: To evaluate the effect of external fixation combined with vacuum sealing drainage on the trauma degree and bone metabolism in patients with open tibiofibula fracture. Methods:A total of 116 patients with op...Objective: To evaluate the effect of external fixation combined with vacuum sealing drainage on the trauma degree and bone metabolism in patients with open tibiofibula fracture. Methods:A total of 116 patients with open tibiofibula fracture who received surgical treatment in Luzhou People's Hospital between February 2015 and January 2017 were divided into control group (n=58) and study group (n=58) by random number table. Control group received debridement + external fixation, and study group received debridement + external fixation +vacuum sealing drainage. The differences in the levels of trauma indexes and bone metabolism indexes were compared between the two groups before and after treatment. Results: Before surgery, there was no statistically significant difference in serum levels of trauma indexes and bone metabolism indexes between the two groups. 1 week after surgery, serum acute phase protein Tf level of study group was higher than that of control group whereas CER, Hp and CRP levels were lower than those of control group;stress indexes NE and Cor levels were lower than those of control group;bone metabolism indexes P1NP, BGP and BALP levels were higher than those of control group whereas β-CTX level was lower than that of control group. Conclusion: External fixation combined with vacuum sealing drainage can effectively reduce fracture trauma and promote fracture end healing in patients with open tibiofibula fracture.展开更多
Objective:To study the effects of placenta polypeptide injection on the hemodynamics and bone metabolism after tibial fracture surgery.Methods:A prospective study was designed,and the patients with tibial fractures wh...Objective:To study the effects of placenta polypeptide injection on the hemodynamics and bone metabolism after tibial fracture surgery.Methods:A prospective study was designed,and the patients with tibial fractures who received surgical treatment in our hospital between July 2015 and September 2017 were selected and randomly divided into the observation group receiving placenta polypeptide injection and the control group receiving conventional therapy.The blood viscosity as well as serum contents of thromboxane B2(TXB2),6-keto-prostaglandin-F1α(6-k-PGF1α)and bone metabolism markers were determined on the day and 14 days after surgery.Results:14 days after treatment,the whole blood high shear viscosity,whole blood low shear viscosity and whole blood middle shear viscosity as well as serum TXB2,C-terminal propeptide of procollagen type I(PICP),N-terminal propeptide of procollagen type I(PINP),osteoprotegerin(OPG),osteocalcin(OC)and alkaline phosphatase(ALP)contents of both groups increased,while serum 6-k-PGF1α,βisomer of C-terminal telopeptide of type I collagen(β-CTX),receptor activator of nuclear factor kB ligand(RANKL)and tartrate-resistant acid phosphatase 5b(TRACP5b)contents decreased,and the whole blood high shear viscosity,whole blood low shear viscosity and whole blood middle shear viscosity as well as serum TXB2,β-CTX,RANKL and TRACP5b contents of the observation group were lower than those of the control group,while serum 6-k-PGF1α,PICP,PINP,OPG,OC and ALP contents were higher than those of the control group.Conclusion:Placenta polypeptide injection therapy after tibial fracture surgery can improve hemodynamics and bone metabolism,which is beneficial to fracture healing.展开更多
Tensors are a popular programming interface for developing artificial intelligence(AI)algorithms.Layout refers to the order of placing tensor data in the memory and will affect performance by affecting data locality;t...Tensors are a popular programming interface for developing artificial intelligence(AI)algorithms.Layout refers to the order of placing tensor data in the memory and will affect performance by affecting data locality;therefore the deep neural network library has a convention on the layout.Since AI applications can use arbitrary layouts,and existing AI systems do not provide programming abstractions to shield the layout conventions of libraries,operator developers need to write a lot of layout-related code,which reduces the efficiency of integrating new libraries or developing new operators.Furthermore,the developer assigns the layout conversion operation to the internal operator to deal with the uncertainty of the input layout,thus losing the opportunity for layout optimization.Based on the idea of polymorphism,we propose a layout-agnostic virtual tensor programming interface,namely the VTensor framework,which enables developers to write new operators without caring about the underlying physical layout of tensors.In addition,the VTensor framework performs global layout inference at runtime to transparently resolve the required layout of virtual tensors,and runtime layout-oriented optimizations to globally minimize the number of layout transformation operations.Experimental results demonstrate that with VTensor,developers can avoid writing layout-dependent code.Compared with TensorFlow,for the 16 operations used in 12 popular networks,VTensor can reduce the lines of code(LOC)of writing a new operation by 47.82%on average,and improve the overall performance by 18.65%on average.展开更多
Frequent itemset mining (FIM) is a popular data mining issue adopted in many fields, such as commodity recommendation in the retail industry, log analysis in web searching, and query recommendation (or related sea...Frequent itemset mining (FIM) is a popular data mining issue adopted in many fields, such as commodity recommendation in the retail industry, log analysis in web searching, and query recommendation (or related search). A large number of FIM algorithms have been proposed to obtain better performance, including parallelized algorithms for processing large data volumes. Besides, incremental FIM algorithms are also proposed to deal with incremental database updates. However, most of these incremental algorithms have low parallelism, causing low efficiency on huge databases. This paper presents two parallel incremental FIM algorithms called IncMiningPFP and IncBuildingPFP, implemented on the MapReduce framework. IncMiningPFP preserves the FP-tree mining results of the original pass, and utilizes them for incremental calculations. In particular, we propose a method to generate a partial FP-tree in the incremental pass, in order to avoid unnecessary mining work. Further, some of the incremental parallel tasks can be omitted when the inserted transactions include fewer items. IncbuildingPFP preserves the CanTrees built in the original pass, and then adds new transactions to them during the incremental passes. Our experimental results show that IncMiningPFP can achieve significant speedup over PFP (Parallel FPGrowth) and a sequential incremental algorithm (CanTree) in most cases of incremental input database, and in other cases IncBuildingPFP can achieve it.展开更多
The machining accuracy of computer numerical control machine tools has always been a focus of the manufacturing industry.Among all errors,thermal error affects the machining accuracy considerably.Because of the signif...The machining accuracy of computer numerical control machine tools has always been a focus of the manufacturing industry.Among all errors,thermal error affects the machining accuracy considerably.Because of the significant impact of Industry 4.0 on machine tools,existing thermal error modeling methods have encountered unprecedented challenges in terms of model complexity and capability of dealing with a large number of time series data.A thermal error modeling method is proposed based on bidirectional long short-term memory(BiLSTM)deep learning,which has good learning ability and a strong capability to handle a large group of dynamic data.A four-layer model framework that includes BiLSTM,a feedforward neural network,and the max pooling is constructed.An elaborately designed algorithm is proposed for better and faster model training.The window length of the input sequence is selected based on the phase space reconstruction of the time series.The model prediction accuracy and model robustness were verified experimentally by three validation tests in which thermal errors predicted by the proposed model were compensated for real workpiece cutting.The average depth variation of the workpiece was reduced from approximately 50μm to less than 2μm after compensation.The reduction in maximum depth variation was more than 85%.The proposed model was proved to be feasible and effective for improving machining accuracy significantly.展开更多
Background:Estrogen is involved in the pathophysiological process of benign prostatic hyperplasia(BPH),in which epithelial-mesenchymal transition(EMT)plays an important role.Upregulation of aquaporin(AQP)5,which is di...Background:Estrogen is involved in the pathophysiological process of benign prostatic hyperplasia(BPH),in which epithelial-mesenchymal transition(EMT)plays an important role.Upregulation of aquaporin(AQP)5,which is directly activated by estrogen,has been reported to promote EMT in multiple cells.This study aimed to examine the effects of AQP5 on estrogen-induced EMT in the prostate.Methods:Normal prostate(NP)tissue samples without any histopathological changes and BPH tissue samples with pathologically confirmed hyperplasia were obtained.An EMT cell model was subsequently established by adding estradiol(E2)to RWPE-1 cells,after which AQP5 knockdown was performed.Tissue morphological and immunohistochemical features were examined using hematoxylin-eosin and immunohistochemical staining.Western blot analysis was performed to determine the expression of AQPs,estrogen receptors,and EMT-related proteins.Cell proliferation was assessed and supernatants were collected for enzyme-linked immunosorbent assay to determine transforming growth factor-β1(TGF-β1)concentrations.Immunofluorescence staining was performed to assess protein expressions in RWPE-1 cells.Results:BPH tissues exhibited greater EMT(TGF-β1:1.362±0.196 vs.0.107±0.067,P=0.003;vimentin:1.581±0.508 vs.0.221±0.047,P<0.001;E-cadherin:0.197±0.188 vs.1.344±0.088,P<0.001),higher AQP5(1.268±0.136 vs.0.227±0.055,P<0.001)and estrogen receptor(ER)α(1.250±0.117 vs.0.329±0.134,P<0.001)expression but lower ERβ(0.271±0.184 vs.1.564±0.130,P<0.001)expression than NP tissues.E2-stimulated cells had higher AQP5 expression(1.298±0.058 vs.1.085±0.104,P=0.049),increased cell proliferation(1.510±0.089 vs.1.000±0.038,P<0.001),and EMT(TGF-β1 concentration:0.352±0.021 ng/mL vs.0.125±0.014 ng/mL,P<0.001;vimentin:1.641±0.120 vs.0.188±0.020,P=0.002;E-cadherin:0.075±0.030 vs.0.843±0.046,P<0.001)than controls.E2-stimulated cells with AQP5 knockdown exhibited decreased EMT(TGF-β1 concentration:0.223±0.041 ng/mL vs.0.352±0.021 ng/mL,P=0.016;vimentin:0.675±0.056 vs.1.641±0.120,P=0.001;E-cadherin:0.159±0.037 vs.0.075±0.030,P=0.040)than E2-stimulated cells with non-related small interfering RNA(siRNA).Conclusion:Our findings suggest that estrogen induces BPH possibly by promoting AQP5 expression.Hence,AQP5 might be a novel target for modulating EMT in prostate epithelial cells.展开更多
In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods ...In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods that rely on circular queues predominantly implemented using indirectly addressable shared memory, our hybrid method exploits a new reuse pattern spanning across the multiple time steps in stencil computations so that circular queues can be implemented by both shared memory and registers effectively in a balanced manner. We describe a framework that automatically finds the best placement of data in registers and shared memory in order to maximize the performance of stencil computations. Validation using four different types of stencils on three different GPU platforms shows that our hybrid method achieves speedups up to 2.93X over methods that use circular queues implemented with shared-memory only.展开更多
GPUs become a ubiquitous choice as coprocessors since they have excellent ability in concurrent processing. In GPU architecture, shared memory plays a very important role in system performance as it can largely improv...GPUs become a ubiquitous choice as coprocessors since they have excellent ability in concurrent processing. In GPU architecture, shared memory plays a very important role in system performance as it can largely improve bandwidth utilization and accelerate memory operations. However, even for affine GPU applications that contain regular access patterns, optimizing for shared memory is not an easy work. It often requires programmer expertise and nontrivial parameter selection. Improper shared memory usage might even underutilize GPU resource: Even using state-of-the-art high level programming models (e.g., OpenACC and OpenHMPP), it is still hard to utilize shared memory since they lack inherent support in describing shared memory optimization and selecting suitable parameters, let alone maintaining high resource utilization. Targeting higher productivity for affine applications, we propose a data centric way to shared memory optimization on GPU. We design a pragma extension on OpenACC so as to convey data management hints of programmers to compiler. Meanwhile, we devise a compiler framework to automatically select optimal parameters for shared arrays, using the polyhedral model. We further propose optimization techniques to expose higher memory and instruction level parallelism. The experimental results show that our shared memory centric approaches effectively improve the performance of five typical GPU applications across four widely used platforms by 3.7x on average, and do not burden programmers with lots of pragmas.展开更多
Increasingly there is a need to process graphs that are larger than the available memory on today's machines.Many systems have been developed with grapli representations that are efficient and compact for out-of-c...Increasingly there is a need to process graphs that are larger than the available memory on today's machines.Many systems have been developed with grapli representations that are efficient and compact for out-of-core processing.A necessary task in these systems is memory management.This paper presents a system called Cacheap which automatically and efficiently manages the available memory to maximize the speed of grapli processing,minimize the amount of disk access,and maximize the utilization of memory for graph data.It has a simple interface that can be easily adopted by existing graph engines.The paper describes the new system,uses it in recent graph engines,and demonstrates its integer factor improvements in the speed of large-scale grapli processing.展开更多
文摘Objective: To evaluate the effect of external fixation combined with vacuum sealing drainage on the trauma degree and bone metabolism in patients with open tibiofibula fracture. Methods:A total of 116 patients with open tibiofibula fracture who received surgical treatment in Luzhou People's Hospital between February 2015 and January 2017 were divided into control group (n=58) and study group (n=58) by random number table. Control group received debridement + external fixation, and study group received debridement + external fixation +vacuum sealing drainage. The differences in the levels of trauma indexes and bone metabolism indexes were compared between the two groups before and after treatment. Results: Before surgery, there was no statistically significant difference in serum levels of trauma indexes and bone metabolism indexes between the two groups. 1 week after surgery, serum acute phase protein Tf level of study group was higher than that of control group whereas CER, Hp and CRP levels were lower than those of control group;stress indexes NE and Cor levels were lower than those of control group;bone metabolism indexes P1NP, BGP and BALP levels were higher than those of control group whereas β-CTX level was lower than that of control group. Conclusion: External fixation combined with vacuum sealing drainage can effectively reduce fracture trauma and promote fracture end healing in patients with open tibiofibula fracture.
基金Science and Technology Planning Project of Wujiaqu of the Sixth Division(No.1537).
文摘Objective:To study the effects of placenta polypeptide injection on the hemodynamics and bone metabolism after tibial fracture surgery.Methods:A prospective study was designed,and the patients with tibial fractures who received surgical treatment in our hospital between July 2015 and September 2017 were selected and randomly divided into the observation group receiving placenta polypeptide injection and the control group receiving conventional therapy.The blood viscosity as well as serum contents of thromboxane B2(TXB2),6-keto-prostaglandin-F1α(6-k-PGF1α)and bone metabolism markers were determined on the day and 14 days after surgery.Results:14 days after treatment,the whole blood high shear viscosity,whole blood low shear viscosity and whole blood middle shear viscosity as well as serum TXB2,C-terminal propeptide of procollagen type I(PICP),N-terminal propeptide of procollagen type I(PINP),osteoprotegerin(OPG),osteocalcin(OC)and alkaline phosphatase(ALP)contents of both groups increased,while serum 6-k-PGF1α,βisomer of C-terminal telopeptide of type I collagen(β-CTX),receptor activator of nuclear factor kB ligand(RANKL)and tartrate-resistant acid phosphatase 5b(TRACP5b)contents decreased,and the whole blood high shear viscosity,whole blood low shear viscosity and whole blood middle shear viscosity as well as serum TXB2,β-CTX,RANKL and TRACP5b contents of the observation group were lower than those of the control group,while serum 6-k-PGF1α,PICP,PINP,OPG,OC and ALP contents were higher than those of the control group.Conclusion:Placenta polypeptide injection therapy after tibial fracture surgery can improve hemodynamics and bone metabolism,which is beneficial to fracture healing.
基金supported by the National Key Research and Development Program of China under Grant No.2021zD0110101the National Natural Science Foundation of China under Grant Nos.62090024,61872043,and 61802368the Australian Research Council grant under Grant Nos.DP180104069 and DP210102409。
文摘Tensors are a popular programming interface for developing artificial intelligence(AI)algorithms.Layout refers to the order of placing tensor data in the memory and will affect performance by affecting data locality;therefore the deep neural network library has a convention on the layout.Since AI applications can use arbitrary layouts,and existing AI systems do not provide programming abstractions to shield the layout conventions of libraries,operator developers need to write a lot of layout-related code,which reduces the efficiency of integrating new libraries or developing new operators.Furthermore,the developer assigns the layout conversion operation to the internal operator to deal with the uncertainty of the input layout,thus losing the opportunity for layout optimization.Based on the idea of polymorphism,we propose a layout-agnostic virtual tensor programming interface,namely the VTensor framework,which enables developers to write new operators without caring about the underlying physical layout of tensors.In addition,the VTensor framework performs global layout inference at runtime to transparently resolve the required layout of virtual tensors,and runtime layout-oriented optimizations to globally minimize the number of layout transformation operations.Experimental results demonstrate that with VTensor,developers can avoid writing layout-dependent code.Compared with TensorFlow,for the 16 operations used in 12 popular networks,VTensor can reduce the lines of code(LOC)of writing a new operation by 47.82%on average,and improve the overall performance by 18.65%on average.
基金This work was supported by the National High Technology Research and Development 863 Program of China under Grant Nos. 2015AA011505, 2015AA015306, and 2012AA010902, the National Natural Science Foundation of China under Grant Nos. 61202055, 61221062, 61521092, 61303053, 61432016, 61402445, and 61672492, and the National Key Research and Development Program of China under Grant No. 2016YFB1000402.
文摘Frequent itemset mining (FIM) is a popular data mining issue adopted in many fields, such as commodity recommendation in the retail industry, log analysis in web searching, and query recommendation (or related search). A large number of FIM algorithms have been proposed to obtain better performance, including parallelized algorithms for processing large data volumes. Besides, incremental FIM algorithms are also proposed to deal with incremental database updates. However, most of these incremental algorithms have low parallelism, causing low efficiency on huge databases. This paper presents two parallel incremental FIM algorithms called IncMiningPFP and IncBuildingPFP, implemented on the MapReduce framework. IncMiningPFP preserves the FP-tree mining results of the original pass, and utilizes them for incremental calculations. In particular, we propose a method to generate a partial FP-tree in the incremental pass, in order to avoid unnecessary mining work. Further, some of the incremental parallel tasks can be omitted when the inserted transactions include fewer items. IncbuildingPFP preserves the CanTrees built in the original pass, and then adds new transactions to them during the incremental passes. Our experimental results show that IncMiningPFP can achieve significant speedup over PFP (Parallel FPGrowth) and a sequential incremental algorithm (CanTree) in most cases of incremental input database, and in other cases IncBuildingPFP can achieve it.
基金sponsored by the National Natural Science Foundation of Major Special Instruments(Grant No.51527806)the National Natural Science Foundation Projects of the People’s Republic of China(Grant No.51975372).
文摘The machining accuracy of computer numerical control machine tools has always been a focus of the manufacturing industry.Among all errors,thermal error affects the machining accuracy considerably.Because of the significant impact of Industry 4.0 on machine tools,existing thermal error modeling methods have encountered unprecedented challenges in terms of model complexity and capability of dealing with a large number of time series data.A thermal error modeling method is proposed based on bidirectional long short-term memory(BiLSTM)deep learning,which has good learning ability and a strong capability to handle a large group of dynamic data.A four-layer model framework that includes BiLSTM,a feedforward neural network,and the max pooling is constructed.An elaborately designed algorithm is proposed for better and faster model training.The window length of the input sequence is selected based on the phase space reconstruction of the time series.The model prediction accuracy and model robustness were verified experimentally by three validation tests in which thermal errors predicted by the proposed model were compensated for real workpiece cutting.The average depth variation of the workpiece was reduced from approximately 50μm to less than 2μm after compensation.The reduction in maximum depth variation was more than 85%.The proposed model was proved to be feasible and effective for improving machining accuracy significantly.
基金This work was supported by grants from the National Key Research and Development Program of China(No.SQ2017YFSF090096)the National Natural Science Foundation of China(Nos.81974098,81770756,81974099,and 81300627)+1 种基金a Special Supportive Program for Organ Transplantation by COTDF(No.2019JYJH08)the Research Funding of Sichuan Health and Family Planning Commission(No.18PJ453)。
文摘Background:Estrogen is involved in the pathophysiological process of benign prostatic hyperplasia(BPH),in which epithelial-mesenchymal transition(EMT)plays an important role.Upregulation of aquaporin(AQP)5,which is directly activated by estrogen,has been reported to promote EMT in multiple cells.This study aimed to examine the effects of AQP5 on estrogen-induced EMT in the prostate.Methods:Normal prostate(NP)tissue samples without any histopathological changes and BPH tissue samples with pathologically confirmed hyperplasia were obtained.An EMT cell model was subsequently established by adding estradiol(E2)to RWPE-1 cells,after which AQP5 knockdown was performed.Tissue morphological and immunohistochemical features were examined using hematoxylin-eosin and immunohistochemical staining.Western blot analysis was performed to determine the expression of AQPs,estrogen receptors,and EMT-related proteins.Cell proliferation was assessed and supernatants were collected for enzyme-linked immunosorbent assay to determine transforming growth factor-β1(TGF-β1)concentrations.Immunofluorescence staining was performed to assess protein expressions in RWPE-1 cells.Results:BPH tissues exhibited greater EMT(TGF-β1:1.362±0.196 vs.0.107±0.067,P=0.003;vimentin:1.581±0.508 vs.0.221±0.047,P<0.001;E-cadherin:0.197±0.188 vs.1.344±0.088,P<0.001),higher AQP5(1.268±0.136 vs.0.227±0.055,P<0.001)and estrogen receptor(ER)α(1.250±0.117 vs.0.329±0.134,P<0.001)expression but lower ERβ(0.271±0.184 vs.1.564±0.130,P<0.001)expression than NP tissues.E2-stimulated cells had higher AQP5 expression(1.298±0.058 vs.1.085±0.104,P=0.049),increased cell proliferation(1.510±0.089 vs.1.000±0.038,P<0.001),and EMT(TGF-β1 concentration:0.352±0.021 ng/mL vs.0.125±0.014 ng/mL,P<0.001;vimentin:1.641±0.120 vs.0.188±0.020,P=0.002;E-cadherin:0.075±0.030 vs.0.843±0.046,P<0.001)than controls.E2-stimulated cells with AQP5 knockdown exhibited decreased EMT(TGF-β1 concentration:0.223±0.041 ng/mL vs.0.352±0.021 ng/mL,P=0.016;vimentin:0.675±0.056 vs.1.641±0.120,P=0.001;E-cadherin:0.159±0.037 vs.0.075±0.030,P=0.040)than E2-stimulated cells with non-related small interfering RNA(siRNA).Conclusion:Our findings suggest that estrogen induces BPH possibly by promoting AQP5 expression.Hence,AQP5 might be a novel target for modulating EMT in prostate epithelial cells.
基金Supported in part by the National Basic Research 973 Program of China under Grant Nos. 2011CB302504 and 2011ZX01028-001-002the National High Technology Research and Development 863 Program of China under Grant No. 2009AA01A129+1 种基金the National Natural Science Foundation of China (NSFC) under Grant No. 60970024the Innovation Research Group of NSFC under Grant No. 60921002
文摘In this paper, we present a hybrid circular queue method that can significantly boost the performance of stencil computations on GPU by carefully balancing usage of registers and shared-memory. Unlike earlier methods that rely on circular queues predominantly implemented using indirectly addressable shared memory, our hybrid method exploits a new reuse pattern spanning across the multiple time steps in stencil computations so that circular queues can be implemented by both shared memory and registers effectively in a balanced manner. We describe a framework that automatically finds the best placement of data in registers and shared memory in order to maximize the performance of stencil computations. Validation using four different types of stencils on three different GPU platforms shows that our hybrid method achieves speedups up to 2.93X over methods that use circular queues implemented with shared-memory only.
基金This work was supported by the National High Technology Research and Development 863 Program of China under Grant No. 2012AA010902, the National Natural Science Foundation of China (NSFC) under Grant No. 61432018, and the Innovation Research Group of NSFC under Grant No. 61221062.
文摘GPUs become a ubiquitous choice as coprocessors since they have excellent ability in concurrent processing. In GPU architecture, shared memory plays a very important role in system performance as it can largely improve bandwidth utilization and accelerate memory operations. However, even for affine GPU applications that contain regular access patterns, optimizing for shared memory is not an easy work. It often requires programmer expertise and nontrivial parameter selection. Improper shared memory usage might even underutilize GPU resource: Even using state-of-the-art high level programming models (e.g., OpenACC and OpenHMPP), it is still hard to utilize shared memory since they lack inherent support in describing shared memory optimization and selecting suitable parameters, let alone maintaining high resource utilization. Targeting higher productivity for affine applications, we propose a data centric way to shared memory optimization on GPU. We design a pragma extension on OpenACC so as to convey data management hints of programmers to compiler. Meanwhile, we devise a compiler framework to automatically select optimal parameters for shared arrays, using the polyhedral model. We further propose optimization techniques to expose higher memory and instruction level parallelism. The experimental results show that our shared memory centric approaches effectively improve the performance of five typical GPU applications across four widely used platforms by 3.7x on average, and do not burden programmers with lots of pragmas.
基金the National Key Research and Development Program of China under Grant No.2017YFB1003103the National Natural Science Foundation of China under Grant Nos.6143201&61432016,61332009,and 61521092+1 种基金the National Science Foundation of USA under Contract Nos.CCF-1717877 and CCF-1629376an IBM CAS Faculty Fellowship.
文摘Increasingly there is a need to process graphs that are larger than the available memory on today's machines.Many systems have been developed with grapli representations that are efficient and compact for out-of-core processing.A necessary task in these systems is memory management.This paper presents a system called Cacheap which automatically and efficiently manages the available memory to maximize the speed of grapli processing,minimize the amount of disk access,and maximize the utilization of memory for graph data.It has a simple interface that can be easily adopted by existing graph engines.The paper describes the new system,uses it in recent graph engines,and demonstrates its integer factor improvements in the speed of large-scale grapli processing.