A From:	SMTP%"RELAY-INFO-VAX@CRVAX.SRI.COM" 12-JUN-1993 11:53:44.34  To:	EVERHART CC:	0 Subj:	Re: Difference between VMSclusters and NFS   X-Newsgroups: comp.os.vms 1 From: zrepachol@cc.curtin.edu.au (Paul Repacholi) 3 Subject: Re: Difference between VMSclusters and NFS 0 Message-Id: <1993Jun8.055749.1@cc.curtin.edu.au>	 Lines: 59 . Sender: news@cujo.curtin.edu.au (News Manager)- Organization: Curtin University of Technology " Date: Mon, 7 Jun 1993 20:57:49 GMT To: Info-VAX@kl.sri.com  X-Gateway-Source-Info: USENET   K In article <1671@se.alcbel.be>, mvbr@se.alcbel.be (Marc Verbruggen) writes: O > Now, my question : In what sense do these 3 technologies differ ? Is the lock Q > manager concept in clusters crucial ? Is there no such thing within Unix or the  > Distributed Services ?  A The distributed lock manager is one of the key bits of a cluster. D The REAL workers though are the MSCP, SCS and the conection manager.B A cluster is very 'real' in that its members co-operate to keep itG intact over time. Hands up all those who have forgoten about a diskless G node in some forgoten corner, and had to do a cluster re-boot again ;-(   I SCS and the conection manager provide a set of comunication services that J MSCP and the DLM can rely on. With SCS and MSCP you can do nearly anythingF NFS can do, including screw up mightly! The problem is, of course, co-I ordination of the use of resorces. So the conection manager maintains the C integrity of the conections, and identifies alternate paths to each H devise. This is essential, even if you don't want the redundancy, as youG must know if your ulimate target is the same devise, or a different one B with a simular name. This is why the cluster manuals rave on about  allocation classes and the like.  G When you have each devise identified, you can then control it use. Note E that clusters use 'discressionary locking', not enforced locking. You A CAN step around the lock manager and access something without any F syncronization to the rest of the cluster, or issue a lock request forI a key resource name and hang the cluster as everone else waits for you to F free it. ( This is the common cause of cluster hangs. Something in oneI machine gets the lock on SYSUAF or the system volume and gets stuck. Then @ all the other machines hang, waiting for the lock to be freed. )H Note, it is not EASY to bypass locking, just possible. For instance, RMSF does all the record and index locking in files. By not useing RMS, andH going straight to the XQP with QIOs you can open the file and access the: data. You just have to do ALL of the data access yourself.  B The lock manager can also be used to comunicate across the clusterE by using the 'lock value block' in the ENQ/DEQ calls. This is the key G ( along with the SCS services ) to things like OPCOM, the que manaager, F license managment, etc. It is also VERY usefull for running a criticalF aplication over a cluster. One copy can create a resource name and putE an exclusive lock on it. All other copies ( on the other nodes ) hang E waiting for the lock. If the active node fails, *ONE* other node will  get the lock and can proceed.   C Infact the DLM had an unexpected plus in non-clustered systems. Pre E VMS v4, all volume managment was done by one process running F11BACP. K This process was responsable for all file level lockinf and syncronisation. E One effect of this was that it was nearly single threaded for all the E system. For v4, it HAD to use the DLM to sync with other processes in H other cluster members, so it could also use the DLM for local sync. Now,F as much of the global state is in the lock manager, there is no reasonE not to let EACH PROCESS independantly manager the file structure, and E use the DLM to sync against ALL other processes in the cluster. Hence D file processing is now multi-threaded! and only has to pause to snycA if there is an access conflict. It also enables each node to pre- C allocate thing and cache them so as to speed things up. If the node J goes down, the inforation in the LVBs can be used to recover and clean up.   ~Paul     