Michael Factor

contact information

IBM Fellow - Hybrid Cloud Data
IBM Research - Israel, Haifa, Israel
+972 dash

829

6350

links

Professional Associations

Professional Associations: ACM | Sigma Xi

2020

RAIDP: replication with intra-disk parity
Eitan Rosenfeld, Aviad Zuck, Nadav Amit, Michael Factor, Dan Tsafrir
Proceedings of the Fifteenth European Conference on Computer Systems, 2020
parity, parallel computing, distributed computing, computer science

2018

Shared Cloud Object Store, governed by permissioned blockchain
Artem Barger, Yacov Manevich, Vita Bortnikov, Yoav Tock, Michael Factor, Michal Malka
Proceedings of the 11th ACM International Systems and Storage Conference, SYSTOR 2018, HAIFA, Israel, June 04-07, 2018, pp. 114

Stocator: providing high performance and fault tolerance for apache spark over object storage
Gil Vernik, Michael Factor, Elliot K. Kolodner, Pietro Michiardi, Effi Ofer, Francesco Pace
Proceedings of the 18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, pp. 462-471, 2018
Abstract speculative execution, spark, service provider, operating system, object storage, fault tolerance, data structure, computer science, computer data storage, big data

2017

Stocator: an object store aware connector for apache spark
Gil Vernik, Michael Factor, Elliot K. Kolodner, Effi Ofer, Pietro Michiardi, Francesco Pace
Proceedings of the 2017 Symposium on Cloud Computing, pp. 653-653
Abstract real time computing, operating system, object storage, method, information repository, eventual consistency, data transfer object, converged storage, computer science, cloud computing, binary large object

Data is the natural resource of the 21st century. It is being produced at dizzying rates, e.g., for genomics, for media and entertainment, and for Internet of Things. Object storage systems such as Amazon S3, Azure Blob storage, and IBM Cloud Object Storage, are highly scalable distributed storage systems that offer high capacity, cost effective storage. But it is not enough just to store data; we also need to derive value from it. Apache Spark is the leading big data analytics processing engine combining MapReduce, SQL, streaming, and complex analytics. We present Stocator, a high performance storage connector, enabling Spark to work directly on data stored in object storage systems, while providing the same correctness guarantees as Hadoops original storage system, HDFS. 122 123 Current object storage connectors from the Hadoop community, e.g., for the S3 and Swift APIs, do not deal well with eventual consistency, which can lead to failure. These connectors assume file system semantics, which is natural given that their model of operation is based on interaction with HDFS. In particular, Spark and Hadoop achieve fault tolerance and enable speculative execution by creating temporary files, listing directories to identify these files, and then renaming them. This paradigm avoids interference between tasks doing the same work and thus writing output with the same name. However, with eventually consistent object storage, a container listing may not yet include a recently created object, and thus an object may not be renamed, leading to incomplete or incorrect results. 247 248 Solutions such as EMRFS [1] from Amazon, S3mper [4] from Netflix, and S3Guard [2], attempt to overcome eventual consistency by requiring additional strongly consistent data storage. These solutions require multiple storage systems, are costly, and can introduce issues of consistency between the stores. 292 293 Current object storage connectors from the Hadoop community are also notorious for their poor performance for write workloads. This, too, stems from their use of the rename operation, which is not a native object storage operation; not only is it not atomic, but it must be implemented using a costly copy operation, followed by delete. Others have tried to improve the performance of object storage connectors by eliminating rename, e.g., the Direct-ParquetOutputCommitter [5] for S3a introduced by Databricks, but have failed to preserve fault tolerance and speculation. 381 382 Stocator takes advantage of object storage semantics to achieve both high performance and fault tolerance. It eliminates the rename paradigm by writing each output object to its final name. The name includes both the part number and the attempt number, so that multiple attempts to write the same part use different objects. Stocator proposes to extend an already existing success indicator object written at the end of a Spark job, to include a manifest with the names of all the objects that compose the final output; this ensures that a subsequent job will correctly read the output, without resorting to a list operation whose results may not be consistent. By leveraging the inherent atomicity of object creation and using a manifest we obtain fault tolerance and enable speculative execution; by avoiding the rename paradigm we greatly decrease the complexity of the connector and the number of operations on the object storage. 534 535 We have implemented our connector and shared it in open source [3]. We have compared its performance with the S3a and Hadoop Swift connectors over a range of workloads and found that it executes many fewer operations on the object storage, in some cases as few as one thirtieth. Since the price for an object storage service typically includes charges based on the number of operations executed, this reduction in operations lowers the costs for clients in addition to reducing the load on client software. It also reduces costs and load for the object storage provider since it can serve more clients with the same amount of processing power. Stocator also substantially increases performance for Spark workloads running over object storage, especially for write intensive workloads, where it is as much as 18 times faster.

Stocator: a high performance object store connector for spark
Gil Vernik, Michael Factor, Elliot K. Kolodner, Effi Ofer, Pietro Michiardi, Francesco Pace
Proceedings of the 10th ACM International Systems and Storage Conference on , 2017
Abstract spark, sql, operating system, object storage, memory management, database, computer science, cloud computing, binary large object, big data, analytics

2016

Using Storage Class Memory Efficiently for an In-memory Database
Gottesman, Yonatan and Nider, Joel and Kat, Ronen and Weinsberg, Yaron and Factor, Michael
Proceedings of the 9th ACM International on Systems and Storage Conference, pp. 21, 2016

2013

Is the Future of Preservation Cloudy
Erik Elmroth, Michael Factor, Ethan L. Miller, Margo I. Seltzer
Dagstuhl Reports 2(11), 102-134, 2013
Abstract world wide web, valuation, software as a service, reuse, obsolescence, forgetting, engineering management, engineering, data preservation, cloud computing, breakout

Is the Future of Preservation Cloudy?(Dagstuhl Seminar 12472)
Elmroth, Erik and Factor, Michael and Miller, Ethan and Seltzer, Margo
Dagstuhl Reports 2(11), Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik, 2013

Secure Logical Isolation for Multi-tenancy in cloud storage
Michael Factor, David Hadas, Aner Hamama, Nadav Har'el, Elliot K Kolodner, Anil Kurmus, Alexandra Shulman-Peleg, Alessandro Sorniotti
Mass Storage Systems and Technologies (MSST), 2013 IEEE 29th Symposium on, pp. 1--5

Cooperative caching with return on investment
Gala Yadgar, Michael Factor, Assaf Schuster
Mass Storage Systems and Technologies (MSST), 2013 IEEE 29th Symposium on, pp. 1--13

2012

Adding advanced storage controller functionality via low-overhead virtualization.
Ben-Yehuda, Muli and Factor, Michael and Rom, Eran and Traeger, Avishay and Borovik, Eran and Yassour, Ben-Ami
FAST, pp. 15--15, 2012

Is the Future of Preservation Cloudy?(Dagstuhl Seminar 12472).
Erik Elmroth, Michael Factor, Ethan Miller, Margo I Seltzer, Erik Elmroth, Michael Factor, Ethan Miller, Margo Seltzer
Dagstuhl Reports 2(11), 102--134, 2012

Adding advanced storage controller functionality via low-overhead virtualization
Muli Ben-Yehuda, Michael Factor, Eran Rom, Avishay Traeger, Eran Borovik, Ben-Ami Yassour
USENIX Conference on File \& Storage Technologies (FAST), 2012

2011

Management of Multilevel, Multiclient Cache Hierarchies with Application Hints
G. Yadgar, M. Factor, K. Li, and A. Schuster
ACM Transactions on Computer Systems 29(2), 2011

2010

The Turtles Project: Design and Implementation of Nested Virtualization
M Ben-Yehuda, M D Day, Z Dubitzky, M Factor, N HarEl, A Gordon, A Liguori, O Wasserman, B A Yassour
OSDI '10: 9th USENIX Symsposium on Operating Systems Design and Implementation, pp. 423--436, 2010

2009

Authenticity and Provenance in Long Term Digital Preservation: Modeling and Implementation in Preservation Aware Storage
Michael Factor, Ealan Henis, Dalit Naor, Simona Rabinovici-Cohen, Petra Reshef, Shahar Ronen, Giovanni Michetti, Maria Guercio
First Workshop on the Theory and Practice of Provenance, February 23, pp. 1--10, 2009

NAP: a building block for remediating performance bottlenecks via black box network analysis
Muli Ben-Yehuda, David Breitgand, Michael Factor, Hillel Kolodner, Valentin Kravtsov, Dan Pelleg
Proceedings of the 6th international conference on Autonomic computing, pp. 179--188, ACM, 2009
Abstract

Optimistic concurrency for clusters via speculative locking
M Factor, A Schuster, K Shagin, T Zamir
Proceedings of SYSTOR 2009: The Israeli Experimental Systems Conference, pp. 1--11

Storage modeling for power estimation
M Allalouf, Y Arbitman, M Factor, R I Kat, K Meth, D Naor
Proceedings of SYSTOR 2009: The Israeli Experimental Systems Conference, pp. 1--10

2008

Preservation of Real-World Data: The Case for Preservation DataStores
S Cohen, M Factor, L Ramati, P Reshef, S Ronen
2008 - storageconference.net

Mc2: Multiple clients on a multilevel cache
G Yadgar, M Factor, K Li, A Schuster
Distributed Computing Systems, 2008, pp. 722--730

Preservation DataStores: New storage paradigm for preservation environments
Simona Rabinovici-Cohen, Michael Factor, Dalit Naor, Leeat Ramati, Petra Reshef, Shahar Ronen, Julian Satran, David L Giaretta
IBM Journal of Research and Development 52(4-5), 389-400, 2008

The ANSI T10 Object-based Storage Standard and Current Implementations
M Factor, S Iren, D Nagle, D Naor, E Riedel, O Rodeh, J Satran
IBM Journal of Research and Development 52(4-5), 401--412, 2008

Virtual machine time travel using continuous data protection and checkpointing
P Ta-Shma, G Laden, M Ben-Yehuda, M Factor
ACM SIGOPS Operating Systems Review 42(1), 127--134, ACM, 2008

2007

Karma: Know-it-all replacement for a multilevel cache
G Yadgar, M Factor, A Schuster
Proceedings of the 5th USENIX conference on File and Storage Technologies, pp. 25--25, 2007

Capability based Secure Access Control to Networked Storage Devices
M Factor, D Naor, E Rom, J Satran, S Tal
Proceedings of the 24th IEEE Conference on Mass Storage Systems and Technologies (MSST 2007), pp. 24--27

Preservation DataStores: Architecture for Preservation Aware Storage
M Factor, D Naor, S Rabinovici
Cohen, L Ramati, P … - Proceedings of the 24th IEEE Conference on Mass Storage …, 2007 - doi.ieeecomputersociety.org

The need for preservation aware storage: a position paper
M Factor, D Naor, S Rabinovici
Cohen, L Ramati, P … - ACM SIGOPS Operating Systems Review, 2007 - portal.acm.org

Architectures for Controller Based CDP
G Laden, P Ta
Shma, E Yaffe, M Factor, S Fienblit - usenix.org, 2007

2006

A platform-independent distributed runtime for standard multithreaded Java
M Factor, A Schuster, K Shagin
International Journal of Parallel Programming 34(2), 113--142, Springer, 2006

2005

Integrating Search and Browse Capabilities into File Systems Extended Abstract-for a Technology …
B Mandler, N Kraus, A Azagury, M Factor
Local to Global Data Interoperability-Challenges and …, 2005 - ieeexplore.ieee.org

The OSD security protocol
M Factor, D Nagle, D Naor, E Riedel, J Satran
Proceedings of the 3rd International IEEE Security in …, 2005 - doi.ieeecomputersociety.org

Object storage: The future building block for storage systems
M. Factor, K. Meth, D. Naor, O. Rodeh, J. Satran
Local to Global Data Interoperability-Challenges and Technologies, 2005, pp. 119--123

2004

Instrumentation of standard libraries in object-oriented languages: the twin class hierarchy approach
M Factor, A Schuster, K Shagin
Proceedings of the 19th annual ACM SIGPLAN conference on Object-oriented programming, systems, languages, and applications, pp. 288--300, 2004

A distributed runtime for Java: yesterday and today
M Factor, A Schuster, K Shagin
18th International Parallel and Distributed Processing Symposium (IPDPS'04) - Workshop 5, 2004

2003

Towards an object store
A. Azagury, V. Dreizin, M. Factor, E. Henis, D. Naor, N. Rinetzky, O. Rodeh, J. Satran, A. Tavory, L. Yerushalmi
Mass Storage Systems and Technologies, 2003.(MSST 2003). Proceedings. 20th IEEE/11th NASA Goddard Conference on, pp. 165--176

JavaSplit: a Runtime for execution of monolithic Java programs on heterogenous collections of commodity workstations
M Factor, A Schuster, K Shagin
Cluster Computing, 2003, pp. 110--117

Advanced functions for storage subsystems: Supporting continuous availability
AC Azagury, ME Factor, WF Micka
IBM SYSTEMS JOURNAL, 2003 - research.ibm.com

2002

A two layered approach for securing an object store network
A Azagury, R Canetti, M Factor, S Halevi, E Henis, D Naor, N Rinetzky, O Rodeh, J Satran
Security in Storage Workshop, 2002, pp. 10--23

Execution of monolithic Java programs on large non-dedicated collections of commodity workstations
M Factor, A Schuster, K Shagin
Proceedings of the 2002 joint ACM-ISCOPE conference on Java Grande, pp. 227--227

A novel navigation paradigm for XML repositories
A Azagury, M E Factor, Y S Maarek, B Mandler
Journal of the American Society for Information Science and Technology 53(6), 515--525, Citeseer, 2002

DYNAMO-DirectorY, Net Archiver and MOver
M Silberstein, M Factor, D Lorenz
International Workshop on Grid Computing, pp. 256--267, Springer, 2002
Abstract http://www.springerlink.com/index/TWCPGF951QQNLALF.pdf

Point-in-time copy: Yesterday, today and tomorrow
A Azagury, ME Factor, J Satran
Proceedings of the Tenth Goddard Conference on Mass Storage …, 2002 - library.gsfc.nasa.gov

2001

A distributed implementation of a virtual machine for java
Y Aridor, M Factor, A Teperman
Concurrency and Computation: practice and experience 13(3), 221--244, John Wiley \& Sons, 2001

Compression in the presence of shared data
M Factor, D Sheinwald
Information Sciences 135(1-2), 29--41, Elsevier, 2001

Implementing Java on Clusters
Y Aridor, M Factor, A Teperman
LECTURE NOTES IN COMPUTER SCIENCE, 2001 - Springer

2000

Speculative field level caching in distributed object-based systems
Y Aridor, M Factor, A Teperman, T Eilam, A …
Technical report, IBM Research Laboratory in Haifa, 2000

XMLFS: an XML-aware file system
A Azagury, M Factor, B Mandler
IBM Research Lab in Haifa, 2000

A high performance cluster JVM presenting a pure single system image
Y Aridor, M Factor, A Teperman, T Eilam, A …
Proceedings of the ACM 2000 conference on Java Grande, 2000 - portal.acm.org

Transparently Obtaining Scalability for Java Applications on a Cluster
Y Aridor, M Factor, A Teperman, T Eilam, A …
Journal of Parallel and Distributed Computing, 2000 - ingentaconnect.com

1999

cJVM: A cluster aware JVM
Y Aridor, M Factor, A Teperman
Proceedings of International Conference on Parallel Processingâ€™99, pp. 31--39, 1999

cJVM: a Single System Image of a JVM on a Cluster
Y Aridor, M Factor, A Teperman
International Conference on Parallel Processing, 1999 - doi.ieeecomputersociety.org

1998

Timing dependent bugs
M Factor, E Farchi, Y Talmor
Software Testing Analysis and Review (STAR98,null), May, 1998

1997

Rigorous testing using SnapShot
M Factor, E Farchi, S Ur
Israeli Conference on Computer Systems and Software Engineering, Herzliya, Israel, pp. 12--21, 1997

1996

Using Snapshot to Implement Hierarchy of Criteria
M Factor, E Farchi, A Gluska, S Ur
International Workshop on Computer Aided Design, Test and Evaluation for Dependabilty, JUL, 1996

Testing concurrent programs: a formal evaluation of coverage criteria
M Factor, E Farchi, Y Lichtenstein, Y Malka
Seventh Israeli Conference on Computer-Based Systems and Software Engineering (ICCSSE '96), pp. 119, 1996

1992

BIO-SPEAD: a parallel computing environment to accelerate development of biologic signal processing …
MI Oppenheim, M Factor, DF Sittig
Comput Methods Programs Biomed, 1992 - ncbi.nlm.nih.gov

The multi-trellis software architecture and the intelligent cardiovascular monitor.
M Factor, DH Gelernter, DF Sittig
Methods Inf Med, 1992 - ncbi.nlm.nih.gov

1991

Tech Rpt. 852, Yale Univ
M Factor
Dept. of Computer Sci., Mar, 1991

Multiple Trellises and the Intelligent Cardiovascular Monitor
M Factor, D Gelernter, DF Sittig
Yale Univ. Dept. ????. Sci. Tech. Report, 1991

Software Backplanes, Realtime Data Fusion and the Process Trellis
M Factor, D Gelernter
Technical Report, 1991

Using Linda to Build Parallel AI Applications, Yale University Department of Computer Science
M Factor, S Fertig, D Gelernter
TR-861, June 1991

The process trellis software architecture for parallel, real-time monitors
ME Factor
1991 - portal.acm.org

A formal specification of the process Trellis
M Factor
PARLE'91 Parallel Architectures and Languages Europe, pp. 277--294, 1991

Using Linda to Build Parallel AI Applications
M Factor, S Fertig, DH Gelernter
Yale University, New Haven,(Jun. 1991) - uta.edu

Real-Time Data Fusion in the Intensive Care Unit
M Factor, DH Gelernter, CE Kolb, PL Miller, DF …
IEEE Computer, 1991 - doi.ieeecs.org

1990

Parallel multi-channel biologic signal processing
DF Sittig, M Factor
Engineering in Medicine and Biology Society, 1990, pp. 1365--1366

Physiologic trend detection and artifact rejection: a parallel implementation of a multi-state Kalman filtering algorithm
D F Sittig, M Factor
Computer Methods and Programs in Biomedicine 31(1), 1--10, Elsevier, 1990

The trellis architecture for intelligent monitors
M Factor, DH Gelernter
5th IEEE International Symposium on Intelligent Control, 1990, pp. 300--303

DYNASCENE: an approach to computer-based intelligent cardiovascular monitoring using sequential …
AI Cohn, S Rosenbaum, M Factor, PL Miller
Methods Inf Med, 1990 - ncbi.nlm.nih.gov

The process trellis architecture for real-time monitors
M Factor
ACM SIGPLAN Notices, 1990 - portal.acm.org

A parallel software architecture for building intelligent medical monitors
M Factor, DF Sittig, AI Cohn, D Gelernter, PL …
Journal of Clinical Monitoring and Computing, 1990 - Springer

1989

Sequential Clinical â€œScenesâ€: A Paradigm for Computer-Based Intelligent Hemodynamic Monitoring
A I Cohn, S Rosenbaum, M Factor, D F Sittig, D Gelernter, P L Miller
Proceedings of the Annual Symposium on Computer Application in Medical Care, pp. 5, 1989

The process trellis: a software architecture for intelligentmonitors
M Factor, DH Gelernter
Tools for Artificial Intelligence, 1989. Architectures, …, 1989 - ieeexplore.ieee.org

A PARALLEL PROCESS LATTICE MODEL FOR AN INTELLIGENT CARDIOVASCULAR MONITOR
PL Miller, M Factor, D Gelernter, DF Slttlg, AI …
AAMSI Congress: Proceedings of the Congress on Medical …, 1989 - The Association

Michael Factor gen_feedback_link(leftr, rightr);