Over the last years, multicores have become accessible to the common developer but writing concurrent programs that are correct and that display good performance still is a hard task. Software Transactional Memory (STM) is a step in the direction of solving the first problem, but it does not provide tools for the programmer to understand and optimize his code's performance, thus leaving the second problem as an open issue. In this paper, we present a novel technique that informs the developer about which objects cause JVSTM transactions to conflict. Then, we describe how we used that technique together with several JVSTM conflict reduction techniques to improve the performance of a transactional application.
{"title":"Profiling and tuning the performance of an STM-based concurrent program","authors":"Luís Pina, João P. Cachopo","doi":"10.1145/2095050.2095069","DOIUrl":"https://doi.org/10.1145/2095050.2095069","url":null,"abstract":"Over the last years, multicores have become accessible to the common developer but writing concurrent programs that are correct and that display good performance still is a hard task. Software Transactional Memory (STM) is a step in the direction of solving the first problem, but it does not provide tools for the programmer to understand and optimize his code's performance, thus leaving the second problem as an open issue.\u0000 In this paper, we present a novel technique that informs the developer about which objects cause JVSTM transactions to conflict. Then, we describe how we used that technique together with several JVSTM conflict reduction techniques to improve the performance of a transactional application.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124519096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
While parallel programming for very regular problems has been used in the scientific community by non-computer-scientists successfully for a few decades now, concurrent programming and solving irregular problems remains hard. Furthermore, we shift from few expert system programmers mastering concurrency for a constrained set of problems to mainstream application developers being required to master concurrency for a wide variety of problems. Consequently, high-level language virtual machine (VM) research faces interesting questions. What are processor design changes that have an impact on the abstractions provided by VMs to provide platform independence? How can application programmers' diverse needs be facilitated to solve concurrent programming problems? We argue that VMs will need to be ready for a wide range of different concurrency models that allow solving concurrency problems with appropriate abstractions. Furthermore, they need to abstract from heterogeneous processor architectures, varying performance characteristics, need to account for memory access cost and inter-core communication mechanisms but should only expose the minimal useful set of notions like locality, explicit communication, and adaptable scheduling to maintain their abstracting nature. Eventually, language designers need to be enabled to guarantee properties like encapsulation, scheduling guarantees, and immutability also when an interaction between different problem-specific concurrency abstractions is required.
{"title":"Which problems does a multi-language virtual machine need to solve in the multicore/manycore era?","authors":"Stefan Marr, M. D. Wael, M. Haupt, T. D'Hondt","doi":"10.1145/2095050.2095104","DOIUrl":"https://doi.org/10.1145/2095050.2095104","url":null,"abstract":"While parallel programming for very regular problems has been used in the scientific community by non-computer-scientists successfully for a few decades now, concurrent programming and solving irregular problems remains hard. Furthermore, we shift from few expert system programmers mastering concurrency for a constrained set of problems to mainstream application developers being required to master concurrency for a wide variety of problems.\u0000 Consequently, high-level language virtual machine (VM) research faces interesting questions. What are processor design changes that have an impact on the abstractions provided by VMs to provide platform independence? How can application programmers' diverse needs be facilitated to solve concurrent programming problems?\u0000 We argue that VMs will need to be ready for a wide range of different concurrency models that allow solving concurrency problems with appropriate abstractions. Furthermore, they need to abstract from heterogeneous processor architectures, varying performance characteristics, need to account for memory access cost and inter-core communication mechanisms but should only expose the minimal useful set of notions like locality, explicit communication, and adaptable scheduling to maintain their abstracting nature.\u0000 Eventually, language designers need to be enabled to guarantee properties like encapsulation, scheduling guarantees, and immutability also when an interaction between different problem-specific concurrency abstractions is required.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127825499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A cornerstone in Domain-Specific Modeling is the definition of modeling languages. A widely used method to formalize domain-specific languages is the metamodeling approach. There are a huge number of metamodeling languages. The choice of a suitable metamodeling approach is a challenging task because there is often a lack of knowledge about the selection criteria and the offered metamodeling features. In this paper, we analyze a set of metamodeling languages (ARIS, Ecore, GME, GOPPRR, MS DSL Tools, and MS Visio), define a comparison framework, and compare the selected meta-metamodels. The comparison forms a first foundation for solving the selection problem.
{"title":"Towards a comparative analysis of meta-metamodels","authors":"Heiko Kern, Axel Hummel, Stefan Kühne","doi":"10.1145/2095050.2095053","DOIUrl":"https://doi.org/10.1145/2095050.2095053","url":null,"abstract":"A cornerstone in Domain-Specific Modeling is the definition of modeling languages. A widely used method to formalize domain-specific languages is the metamodeling approach. There are a huge number of metamodeling languages. The choice of a suitable metamodeling approach is a challenging task because there is often a lack of knowledge about the selection criteria and the offered metamodeling features. In this paper, we analyze a set of metamodeling languages (ARIS, Ecore, GME, GOPPRR, MS DSL Tools, and MS Visio), define a comparison framework, and compare the selected meta-metamodels. The comparison forms a first foundation for solving the selection problem.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132460074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes experiences of the instructor of a course dealing with mobile phone programming. This instance of the course (offered yearly since 2010) reuses the academic content of a traditional software engineering course, but requires mobile phone application development for concrete deliverables that exemplify competency of the academic concepts of the course. The paper describes the tradeoffs between teaching the material vs. students learning the material, group dynamics and constraints, as well as technical recommendations for faculty who are considering offering a course that concentrates on mobile phone applications.
{"title":"Teaching students to learn to learn mobile phone programming","authors":"J. Sprinkle","doi":"10.1145/2095050.2095094","DOIUrl":"https://doi.org/10.1145/2095050.2095094","url":null,"abstract":"This paper describes experiences of the instructor of a course dealing with mobile phone programming. This instance of the course (offered yearly since 2010) reuses the academic content of a traditional software engineering course, but requires mobile phone application development for concrete deliverables that exemplify competency of the academic concepts of the course. The paper describes the tradeoffs between teaching the material vs. students learning the material, group dynamics and constraints, as well as technical recommendations for faculty who are considering offering a course that concentrates on mobile phone applications.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130577105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Domain-Specific Languages (DSLs) allow modelling concerns at a high abstraction level. This simplifies the modelling process and ensures that non-technical stakeholders can be more closely involved in software development. However, increasing the abstraction level causes details of the problem domain to be excluded from the problem space. In some situations, this may render a DSL useless since required details can not be captured by the language. In this paper we explore how generic metamodels can be parameterised to model additional details and thereby increase the reuse value of DSLs.
{"title":"Advancing generic metamodels","authors":"H. Berg, B. Møller-Pedersen, Stein Krogdahl","doi":"10.1145/2095050.2095055","DOIUrl":"https://doi.org/10.1145/2095050.2095055","url":null,"abstract":"Domain-Specific Languages (DSLs) allow modelling concerns at a high abstraction level. This simplifies the modelling process and ensures that non-technical stakeholders can be more closely involved in software development. However, increasing the abstraction level causes details of the problem domain to be excluded from the problem space. In some situations, this may render a DSL useless since required details can not be captured by the language. In this paper we explore how generic metamodels can be parameterised to model additional details and thereby increase the reuse value of DSLs.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2011-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134563131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sequential programs are often difficult to parallelize because of the complexity in their implementation and the uncertainty in their behavior. We will demonstrate behavior-oriented parallelization (BOP), which provides annotations for a user to mark possibly parallel regions (PPR) in code and a safe implementation to execute the annotated tasks in parallel as long as they produce the correct result - the same output as the unmarked sequential execution. This position paper gives a description of the interface and implementation and shows two examples.
{"title":"Parallel programming by hints","authors":"C. Ding, Zachary Fletcher","doi":"10.1145/2095050.2095068","DOIUrl":"https://doi.org/10.1145/2095050.2095068","url":null,"abstract":"Sequential programs are often difficult to parallelize because of the complexity in their implementation and the uncertainty in their behavior. We will demonstrate behavior-oriented parallelization (BOP), which provides annotations for a user to mark possibly parallel regions (PPR) in code and a safe implementation to execute the annotated tasks in parallel as long as they produce the correct result - the same output as the unmarked sequential execution. This position paper gives a description of the interface and implementation and shows two examples.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133711408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Juha-Pekka Tolvanen, J. Sprinkle, Jeffrey G. Gray, M. Rossi
Domain-Specific Modeling raises the level of abstraction beyond programming by specifying the solution directly using visual models to express domain concepts. In many cases, final products can be generated automatically from these high-level specifications. This automation is possible because both the language and generators fit the requirements of only one domain. This paper introduces Domain-Specific Modeling and describes the related 2-day workshop DSM'11.
{"title":"The 11th workshop on domain-specific modeling","authors":"Juha-Pekka Tolvanen, J. Sprinkle, Jeffrey G. Gray, M. Rossi","doi":"10.1145/2095050.2095052","DOIUrl":"https://doi.org/10.1145/2095050.2095052","url":null,"abstract":"Domain-Specific Modeling raises the level of abstraction beyond programming by specifying the solution directly using visual models to express domain concepts. In many cases, final products can be generated automatically from these high-level specifications. This automation is possible because both the language and generators fit the requirements of only one domain. This paper introduces Domain-Specific Modeling and describes the related 2-day workshop DSM'11.","PeriodicalId":143880,"journal":{"name":"SPLASH Workshops","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126569913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：