greg/stasis-aries-wal
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

bibliography entries

Browse source
This commit is contained in:
Sears Russell 2006-04-25 02:38:55 +00:00
parent 7643cb6275
commit a426824f18
2 changed files with 238 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

223
doc/paper3/LLADD.bib
View file

@ -87,6 +87,33 @@
}
@Misc{reiserfs,
OPTkey = {},
author = {Hans T. Reiser},
title = {{ReiserFS}},
OPThowpublished = {},
OPTmonth = {},
OPTyear = {},
note = {http://www.namesys.com},
OPTannote = {}
}
@Misc{sqlserver,
OPTkey = {},
OPTauthor = {},
title = {Microsoft {SQL S}erver 2005},
OPThowpublished = {},
OPTmonth = {},
OPTyear = {},
OPTnote = {},
OPTannote = {}
}
@Article{lrvm,
author = {M. Satyanarayanan and Henry H. Mashburn and Puneet Kumar and David C. Steere and James J. Kistler},
title = {Lightweight recoverable virtual memory},
@ -214,3 +241,199 @@
OPTannote = {}
}
@InProceedings{libtp,
author = {Margo Seltzer and M Olsen},
title = {{LIBTP}: Portable, Modular Transactions for {UNIX}},
OPTcrossref = {},
OPTkey = {},
booktitle = {Proceedings of the 1992 Usenix Conference},
OPTpages = {},
year = {1992},
OPTeditor = {},
OPTvolume = {},
OPTnumber = {},
OPTseries = {},
OPTaddress = {},
month = {Janurary},
OPTorganization = {},
OPTpublisher = {},
OPTnote = {},
OPTannote = {}
}
@InProceedings{riscDB,
author = {Surajit Chaudhuri and Gerhard Weikum},
title = {Rethinking Database System Architecture: Towards a Self-tuning RISC-style Database System},
OPTcrossref = {},
OPTkey = {},
booktitle = {Proceedings of the 26th International Conference on Very Large Databases},
OPTpages = {},
year = {2000},
OPTeditor = {},
OPTvolume = {},
OPTnumber = {},
OPTseries = {},
OPTaddress = {},
OPTmonth = {},
OPTorganization = {},
OPTpublisher = {},
OPTnote = {},
OPTannote = {}
}
@Article{optimisticConcurrencyControl,
author = {H. T. Kung and John T. Robinson},
title = {On optimistic methods for concurrency control},
journal = {ACM Transactions on Database Systems},
year = {1981},
OPTkey = {},
volume = {6},
number = {2},
pages = {213-226},
month = {June},
OPTnote = {},
OPTannote = {}
}
@Article{optimisticConcurrencyPerformance,
author = {R Agrawal and M J Carey and M Livny},
title = {Concurrency control performance modeling: alternatives and implications},
journal = {ACM Transactions on Database Systems},
year = {1987},
OPTkey = {},
OPTvolume = {},
OPTnumber = {},
OPTpages = {},
OPTmonth = {},
OPTnote = {},
OPTannote = {}
}
@inproceedings{oo7,
author = {Michael J. Carey and
David J. DeWitt and
Jeffrey F. Naughton},
OPTeditor = {Peter Buneman and
Sushil Jajodia},
title = {The oo7 Benchmark},
booktitle = {Proceedings of the 1993 ACM SIGMOD International Conference on
Management of Data},
OPTpublisher = {ACM Press},
year = {1993},
pages = {12-21},
OPTee = {http://doi.acm.org/10.1145/170035.170041, db/conf/sigmod/CareyDN93.html},
OPTcrossref = {DBLP:conf/sigmod/93},
OPTbibsource = {DBLP, http://dblp.uni-trier.de}
}
@InProceedings{lht,
author = {Witold Litwin},
title = {Linear Hashing: A New Tool for File and Table Addressing },
OPTcrossref = {},
OPTkey = {},
booktitle = {Proceedings of the 6th Conference on Very Large Dabatases},
pages = {224-232},
year = {1980},
OPTeditor = {},
OPTvolume = {},
OPTnumber = {},
OPTseries = {},
OPTaddress = {},
OPTmonth = {},
OPTorganization = {},
OPTpublisher = {},
OPTnote = {},
OPTannote = {}
}
@inproceedings{esm,
author = {Michael J. Carey and
David J. DeWitt and
Joel E. Richardson and
Eugene J. Shekita},
OPTeditor = {Wesley W. Chu and
Georges Gardarin and
Setsuo Ohsuga and
Yahiko Kambayashi},
title = {Object and File Management in the {EXODUS} Extensible Database
System},
booktitle = {VLDB'86 Twelfth International Conference on Very Large Data Bases},
OPTfoo = { August 25-28, 1986, Kyoto, Japan, Proceedings},
OPTpublisher = {Morgan Kaufmann},
year = {1986},
OPTisbn = {0-934613-18-4},
pages = {91-100},
OPTee = {db/conf/vldb/CareyDRS86.html},
OPTcrossref = {DBLP:conf/vldb/86},
OPTbibsource = {DBLP, http://dblp.uni-trier.de}
}
@Article{ffs,
author = {M K Mckusick and W N Joy and S J Leffler and R S Fabry},
title = {A Fast File System for {UNIX}},
journal = {ACM Transactions on Computer Systems},
year = {1984},
OPTkey = {},
OPTvolume = {},
OPTnumber = {},
OPTpages = {},
OPTmonth = {},
OPTnote = {},
OPTannote = {}
}
@InProceedings{gfs,
author = {S Ghemawat and H Gobioff and ST Leung},
title = {The {G}oogle file system},
OPTcrossref = {},
OPTkey = {},
booktitle = {Proceedings of the 19th ACM Symposium on Operating Systems Principles},
pages = {29-43},
year = {2003},
OPTeditor = {},
OPTvolume = {},
OPTnumber = {},
OPTseries = {},
OPTaddress = {},
OPTmonth = {},
OPTorganization = {},
OPTpublisher = {},
OPTnote = {},
OPTannote = {}
}
@InProceedings{lfs,
author = {The Design and Implementation of a Log-Structured File System},
title = {Mendel Rosenblum and John K. Ousterhout},
OPTcrossref = {},
OPTkey = {},
booktitle = {Proceedings of the 13th ACM Symposium on Operating Systems Principles},
OPTpages = {},
year = {1992},
OPTeditor = {},
OPTvolume = {},
OPTnumber = {},
OPTseries = {},
OPTaddress = {},
OPTmonth = {},
OPTorganization = {},
OPTpublisher = {},
OPTnote = {},
OPTannote = {}
}
@article{hoard,
author = {Emery D. Berger and Kathryn S. McKinley and Robert D. Blumofe and Paul R. Wilson},
title = {Hoard: a scalable memory allocator for multithreaded applications},
journal = {ACM SIGPLAN Notices},
volume = {35},
number = {11},
year = {2000},
OPTissn = {0362-1340},
pages = {117--128},
OPTdoi = {http://doi.acm.org/10.1145/356989.357000},
OPTpublisher = {ACM Press},
OPTaddress = {New York, NY, USA},
}

29
doc/paper3/LLADD.tex
View file

@ -339,7 +339,7 @@ databases. At its core, it provides the physical database
%It is based on the
%observation that the storage subsystem is a more general (and less
%abstract) component than a monolithic database, and provides a
%standalone implementation of the storage primitives built into
%stand-alone implementation of the storage primitives built into
%most relational database systems~\cite{libtp}.
In particular,
it provides fully transactional (ACID) operations over B-Trees,
@ -411,13 +411,13 @@ to build a system that enables a wider range of data management options.
%e-mail use databases to store unstructured binary data, if at all.
%More recently, WinFS, Microsoft's database based
%file metadata management system, has been replaced in favor of an
%file meta data management system, has been replaced in favor of an
%embedded indexing engine that imposes less structure (and provides
%fewer consistency guarantees) than the original
%proposal~\cite{needtocitesomething}.
%Scaling to the very large doesn't work (SAP used DB2 as a hash table
%for years), search engines, cad/vlsi didn't happen. scalable GIS
%for years), search engines, cad/VLSI didn't happen. scalable GIS
%systems use shredded blobs (terraserver, google maps), scaling to many
%was more difficult than implementing from scratch (winfs), scaling
%down doesn't work (variance in performance, footprint),
@ -703,7 +703,7 @@ each data structure until the end of the transaction. Releasing the
lock after the modification, but before the end of the transaction,
increases concurrency. However, it means that follow-on transactions that use
that data may need to abort if a current transaction aborts ({\em
cascading aborts}). These issues are studied in great detail in terms of optimistic concurrency control~\cite{optimisticConcurrencyControl, optimisticConcurrenctPerformance}.
cascading aborts}). Related issues are studied in great detail in terms of optimistic concurrency control~\cite{optimisticConcurrencyControl, optimisticConcurrencyPerformance}.
Unfortunately, the long locks held by total isolation cause bottlenecks when applied to key
data structures.
@ -735,7 +735,7 @@ implements nested top actions. The extension may be used as follows:
\noindent If the transaction that encloses the operation aborts, the logical
undo will {\em compensate} for its effects, leaving the structural
changes intact.
% Note that this recipe does not ensure iso transactional
% Note that this recipe does not ensure ISO transactional
%consistency and is largely orthogonal to the use of a lock manager.
We have found that it is easy to protect operations that make
@ -790,7 +790,7 @@ properly.
We call such pages ``LSN-free'' pages. Although this technique is
novel for databases, it resembles the mechanism used by
RVM~\cite{rvm}; \yad generalizes the concept and allows it to
RVM~\cite{lrvm}; \yad generalizes the concept and allows it to
co-exist with traditional pages. Furthermore, efficient recovery and
log truncation require only minor modifications to our recovery
algorithm. In practice, this is implemented by providing a buffer manager callback
@ -912,6 +912,7 @@ We discuss existing approaches to the systems presented here when
appropriate.
\subsection{Adding log operations}
\label{sec:wal}
\begin{figure}
\includegraphics[%
width=1\columnwidth]{figs/structure.pdf}
@ -1258,7 +1259,7 @@ we see that update/flush indeed improves memory utilization.
\subsection{Manipulation of logical log entries}
\label{sec:logging}
\begin{figure}
\includegraphics[width=1\columnwidth]{figs/graph-traversal.pdf}
\vspace{-24pt}
@ -1305,7 +1306,7 @@ balancing can be implemented by splitting requests across many nodes.
Similarly, a node can easily service streams of requests from multiple
nodes by combining them into a single log, and processing the log
using operation implementations. For example, this type of optimization
is used by RVM's log-merging operations~\cite{rvm}.
is used by RVM's log-merging operations~\cite{lrvm}.
Furthermore, application-specific
procedures that are analogous to standard relational algebra methods
@ -1421,7 +1422,7 @@ objects~\cite{esm}.
Finally, RVM, recoverable virtual memory, made use of LSN-free pages
so that it could use mmap() to map portions of the page file into
application memory\cite{rvm}. However, without support for logical log entries
application memory\cite{lrvm}. However, without support for logical log entries
and nested top actions, it would be difficult to implement a
concurrent, durable data structure using RVM. We plan to add RVM
style transactional memory to \yad in a way that is compatible with
@ -1438,7 +1439,7 @@ incorporate into \yad.
Different large object storage systems provide different API's.
Some allow arbitrary insertion and deletion of bytes~\cite{esm} or
pages~\cite{sqlserver} within the object, while typical filesystems
provide append-only storage allocation~\cite{ffs,ntfs}.
provide append-only storage allocation~\cite{ffs}.
Record-oriented file systems are an older, but still-used
alternative~\cite{vmsFiles11,gfs}. Each of these API's addresses
different workloads.
@ -1458,8 +1459,8 @@ information about memory management.~\cite{xxx} \rcs{Eric, do you have
a reference for this?}
Finally, many systems take a hybrid approach to allocation. Examples include
databases with blob support\cite{something}, and a number of
filesystems~\cite{reiserfs3,didFFSdoThis}.
databases with blob support, and a number of
filesystems~\cite{reiserfs,ffs}.
We are interested in allowing applications to store records in
the transaction log. Assuming log fragmentation is kept to a
@ -1477,8 +1478,8 @@ complement to such a system, especially given \yads focus on
intelligence and optimizations within a single node, and Boxwood's
focus on multiple node systems. In particular, it would be
interesting to explore extensions to the Boxwood approach that make
use of \yads customizable semantics (Section~\ref{wal}), and fully logical logging
mechanism. (Section~\ref{logging})
use of \yads customizable semantics (Section~\ref{sec:wal}), and fully logical logging
mechanism. (Section~\ref{sec:logging})
\section{Future Work}