ISMM Research Papers
Welcome to the 2017 ACM SIGPLAN International Symposium on Memory Management (ISMM).
Call for Papers
The International Symposium on Memory Management (ISMM 2017) is soliciting full-length submissions covering new work on memory management topics, as well as papers presenting confirmations or refutations of important prior results. Surveys and comparative analyses that shed new light on previously published techniques are also welcome.
Areas of interest include but are not limited to:
- Memory allocation and de-allocation
- Garbage Collection algorithms and implementations
- Formal analysis and verification of memory management algorithms
- Compiler analyses to aid memory management
- Tools to analyze memory usage of programs
- Empirical analysis of memory intensive programs
- Formal analysis and verification of memory intensive programs
- Memory system design and analysis
- Hardware support for memory management
- Memory management for new or non-traditional systems
- Novel memory architectures
- Memory management at datacenter and cloud scales
All papers must be submitted on-line in Portable Document Format (PDF). The submission site is at http://ismm17.hotcrp.com
The Program Committee (PC) and External Review Committee (ERC) will read submissions and judge them on scientific merit, innovation, readability, and relevance. Papers previously published or already being reviewed by another conference are not eligible. If a closely related paper has been submitted elsewhere, the authors must notify the Program Chair as per the SIGPLAN republication policy. Papers should be self-contained. Reviewers are not required to read appendices, so a paper should be intelligible without them. Supplemental or auxiliary materials are not accepted.
Submissions must be in English, formatted to print on US Letter (8.5 x 11 inches) paper, and no more than 10 pages (not including bibliography, excluding well marked appendices) in standard ACM SIGPLAN conference format: two columns, nine-point font (or larger) on a ten-point baseline (or larger), with columns 20pc (3.33in) wide and 54pc (9in) tall, and a column gutter of 2pc (0.33in). Detailed formatting guidelines are available at http://www.sigplan.org/Resources/Author/
Papers that violate the above guidelines will be rejected by the Program Chair.
All accepted papers will appear in the published proceedings.
The ISMM paper reviewing process is double blind: authors are not known to the reviewers and reviewers are not known to authors. This is to authors’ benefit, since research indicates that author anonymity reduces bias in reviewing. Authors are required to make reasonable efforts not to disclose their identities to reviewers. For example, authors should not attach names nor mention institution, research group, project name, etc. Authors should refer to projects and prior work in the third person, e.g. “the XYZ project” or work by “Author et al.”, as with discussion of all related work. Where necessary for flow, authors may add a footnote that a name is withheld for review. Paper drafts should not be made public to reduce the possibility of inadvertently revealing author identity to reviewers. Reviewers are honor-bound not to try to discover authors’ identities, which will be known only by the Program chair until the appropriate point in the Program Committee’s deliberations. Authors are required to declare known conflicts with members of the PC and ERC at submission time, as defined by SIGPLAN guidelines: http://www.sigplan.org/Resources/Policies/Review/
ISMM 2017 follows the practice introduced at earlier instances of ISMM of using a separate External Review Committee as part of the reviewing process. The ERC complements the Program Committee by providing a broader and deeper pool from which to draw expert reviews. The same reviewing standards apply to the ERC as for the PC. However, ERC members review fewer papers and do not participate in the PC meeting. ERC members review and decide acceptance for submissions by PC members. This approach should be more practical with double-blind reviewing than ad hoc expert review assignments as used by a number of conferences. The formal selection process, transparency of its constituency, and the fact that each reviewer will review multiple papers should increase the quality and accountability of reviews as compared to traditional ad hoc expert review assignments.
The rebuttal process will occur end of March 2017 (see Important Dates), and will give the authors opportunity to respond to factual errors in reviews before the Program Committee meets to make its decisions. The committee may, but need not, respond to rebuttals or revise reviews at or after the committee meeting.
Sun 18 JunDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
08:30 - 08:45
08:45 - 09:30
|Bridging the Gap between Memory Performance and Massive Parallelism: The Critical Role of Programming Systems Innovations|
K: Xipeng Shen North Carolina State University
09:30 - 10:30
|NG2C: Pretenuring Garbage Collection with Dynamic Generations for HotSpot Big Data Applications|
|Type-Assisted Automatic Garbage Collection for Lock-Free Data Structures|
11:00 - 12:30
|A Marshalled Data Format for Pointers in Relocatable Data Blocks|
|Flexible and Efficient Memory Object Metadata|
|Shadow State Encoding for Efficient Monitoring of Block-level Properties|
14:00 - 15:30
|Analyzing Memory Management Methods on Integrated CPU-GPU Systems|
|Continuous Checkpointing of HTM Transactions in NVM|
|RTHMS: A Tool for Data Placement on Hybrid Memory System|
16:00 - 17:30
|"What's in a Name?" Going Beyond Allocation Site Names in Heap Analysis|
|A Refinement Hierarchy for Free List Memory Allocators|
|Avoiding Consistency Exceptions Under Strong Memory Consistency Models|