08 Tornado

50 %
50 %
Information about 08 Tornado

Published on October 5, 2007

Author: Sudiksha

Source: authorstream.com

Tornado: Minimizing Locality and Concurrency in a SMP OS:  Tornado: Minimizing Locality and Concurrency in a SMP OS Slide2:  http://www.eecg.toronto.edu/~okrieg/tmp.pdf Why locality matters::  Why locality matters: Faster processors and more complex controllers -> higher memory latencies Write sharing costs Large secondary caches Large cache lines -> false sharing NUMA effects … Goal:  Goal Minimize read/write and write sharing; -> minimize cache coherence overheads Minimize false sharing Minimize distance between accessing processor and target memory module Do real systems do this?:  Do real systems do this? Yes and no Tornado -> adopt design principles to maximize locality and concurrency Map locality and independency which exists in the OS requests from applications into locality and independence in servicing these requests in the kernel or system servers Approach – re-think who data structures are organized and how operations on them are applied Counter ilustration:  Counter ilustration Shared counter, array counter, padded counter Tornado basics:  Tornado basics Individual resources in individual objects Mechanisms: Clustering objects Protected procedure calls Semi-automatic garbage collection / efficient locking Clustered objects:  Clustered objects Appear as a single object Multiple “reps” assigned to handle object references from one (or more) processors Object = granularity of access Operations, synchronization can be applied only to relevant pieces Will make global policies more difficult (e.g., global paging policy) Implementation should reflect object use Cluster Objects Implementation:  Cluster Objects Implementation Mix of replication and partitioning techniques: Process Obj replicated, Regions distributed and created on demand… Combination of object migration, home rep, and other techniques (think distributed shared memory…) Translation tables to handle implementation Per processor to access local reps Global partitioned table across processors to find rep for given object Default “miss” handler May be quite large, but sparse -> let caching mechanisms help keep around only relevant pieces… Dynamic Memory Allocation:  Dynamic Memory Allocation Local allocation – per “node” For small, less than cache-line data, use separate pool Addresses false sharing issue Avoid interrupt disabling by using efficient locks Protected procedure calls:  Protected procedure calls Jumps into address space of a (server) object Microkernel design Client requests serviced on local processors (translation table) Handoff scheduling # server threads == # client threads Stub generator to generate code based on public interface Reference checking Special MetaPort to handle first use of a PPC Parameter passing Mix of registers, mapped stack or memory regions Cross-processor IPC Optimize so that caller spins in trap Synchronization:  Synchronization They separate locking (for updates) & existence guarantees (deallocations) Encapsulate lock within object (better rep), avoid global locks Avoids contention, limits cache coherence operations on lock access Use spin-then-block locks Garbage Collection:  Garbage Collection Essentially RCU Must ensure all persistent and temporary object references are removed Object/rep keeps track of requests made out to it – counter decremented on completion – so when counter is zero no temp references Since first use of object goes through translation table, can determine which processors have object reps, and can use a token scheme to ensure object ref counter is zero for each processor Finally – safe to dealloc object Evaluation:  Evaluation Use of NUMAchine and simulator NUMAchine – ring of 4 stations, each with 4 processors and a memory module, direct mapped caches Simulator different interconnect and cache coherence protocol First validate simulator is OK then use simulator to gather other data:  First validate simulator is OK then use simulator to gather other data Effects of cluster objects:  Effects of cluster objects Page faults frequent, region deletions aren’t Slide17:  NUMAchine, SimOS and SimOS w/ 4-way assoc cache Compared to other arch/OS, MT and MP mode:  Compared to other arch/OS, MT and MP mode MT MP pagefault fstat thread

Add a comment

Related presentations

Related pages

Tornados weltweit - Thomas Sävert Naturgewalten

Drei Tote bei Tornado in Nordfrankreich(Tagesschau, 04.08.08) Tornado hinterlässt in Frankreich Schneise der Verwüstung Fotodokumentation von Bjoern Stumpf
Read more

04.08.2010 Tornado im Mercedes-Benz Museum Stuttgart - YouTube

04.08.2010, 20 Uhr: Vorführung des Rauchabzug-Tornados im Mercedes-Benz-Museum Stuttgart. Mit ca. 34 m Höhe der grösste künstliche Tornado ...
Read more

Tornadoliste Deutschland

Zeitpunkt: 19:08 Uhr MESZ. Von Born aus wurde der Tornado auf dem Borner Bülten beobachtet und fotografiert. Er löste sich nach kurzer Zeit auf, ...
Read more

SC Tornado Westig 08 - fussball.de

Alle Mannschaften, Tabellen und Ergebnisse des Fußballvereins SC Tornado Westig 08 auf einen Blick.
Read more

10.08.2014, Tornado in der Wetterau - YouTube

Der Tornado der Kategorie F1 zog am 10.08.2014 gegen 20:25 Uhr im Wetteraukreis zwischen Wölfersheim und Echzell, genauer gesagt vom ...
Read more

Wetternews: Wo sind Tornados möglich? - Alle Infos zur ...

Tornado - Spezial Wo sind Tornados möglich? Alle Infos zur aktuellen Lage. In den folgenden Karten ist der "Convective Outlook" für den aktuellen Tag ...
Read more

SC Tornado Westig 08 | Facebook

SC Tornado Westig 08. 340 likes · 32 talking about this. Fussballclub in Hemer Westig
Read more

Wetter aktuell: Tornado fegte mit bis zu 250 ...

Auch bei Konstanz soll es zu einem Tornado gekommen sein. Wie die Polizei am Donnerstag berichtete, ... Donnerstag, 14. Mai, 08.45 Uhr: ...
Read more

08.06.2008-Superzelle mit Tornado bei Eggerszell

Wetterlage am 08.06.2008: Am Rande hohen Luftdrucks über Skandinavien gelangte trockenwarme Festlandsluft in den Norden und Osten Deutschlands.
Read more