New answers tagged

0

Michael Brown already answered it, just wanted to add: CSV is currently the best practice storage configuration for both Hyper-V, SQL FCI, and SoFS. 99% of the clustered systems I configure for the customers are built on CSV and remaining 1% uses DAG and AlwaysOn AG like BaronSamedi1958 mentioned above. Also,CSV now has a benefit of ODX (Offloaded Data ...


2

This link lists the benefits of CSVs https://technet.microsoft.com/en-us/library/dd759255(v=ws.11).aspx q1) can multiple VMs on Node1 run in the same clusterdisk? I don't think that is possible (the article says no). Even if you could add multiple VMs to the same Clusterdisk it would mean if a single VM failed or needed to be moved it couldn't be moved ...


1

CSV is layered on top of some physical shared storage so you can have CSV without shared storage. https://en.wikipedia.org/wiki/Cluster_Shared_Volumes You can still have a cluster without shared storage / CSV say Exchange DAG and SQL Server AAG. https://technet.microsoft.com/en-us/library/dd979799(v=exchg.150).aspx ...


0

Set up two partitions, one for GPU, one for CPU jobs. Use MaxCPUsPerNode for each. Set up the nodes using the CPUs Parameter. All MaxCPUsPerNode added should be less or qual than this (available CPUs/Cores/Threads) Use SelectTypeParameters=CR_CPU Use SchedulerType=sched/backfill


0

Some related information can be found in this ServerFault question. Making a cluster out of 2 clusters is unusual (aka a bad design), and making a 4-peer cluster with two peers per location is wasteful and useless (no value above one peer per location). In fact, with a 2-peer per location cluster you risk dual-active contention as the active peer at each ...


0

I faced a situation like yours some months ago, and I solved my problem (synchronization of www files between many nodes belonging to apache user) with btsync, that is a program that keeps many files aligned between two or more machines using BitTorrent protocol. It's secure, uses cryptography, and IMHO is very efficient and transparent. It's worth a try. ...


0

In order to run an application using a spark context it is first necessary to run a Slurm job which starts a master and some workers. There are some things you will have to watch out for when using Slurm: don't start Spark as a daemon make the Spark workers use only as much cores and memory as requested for the Slurm job in order to run master and worker ...


4

You have two options in order to create a Hyper-V Failover Cluster using your current hardware: 1) Windows Server 2016 with Storage Replica http://www.tech-coffee.net/storage-replica/ that assumes a manual failover in case one of the servers fails or Storage Spaces Direct http://windowsitpro.com/windows-server/what-storage-spaces-direct that need at ...


0

Use DNS Round robin if you have more than one IP address, Otherwise I would encourage you to use Pacemaker+Corosync with IP and Apache failover How To Set Up an Apache Active-Passive Cluster Using Pacemaker on CentOS 7


2

I don't have a solution for you but I kinda felt this deserves at least some sort of answer: To state the obvious, a machine is highly dependent on its storage. In a highly-available environment - where all VMs can run on any node - the nodes must all be able to see the storage for a VM otherwise they cannot be a host for that VM. So clustered shared ...


0

There is a read preference value "nearest" which actually uses the latency between the MongoS and MongoD to determine which is the best/fastest set member for the query. You can find the documentation about it here: https://docs.mongodb.org/manual/reference/read-preference/#nearest The selection is based on the "member selection". For a sharded cluster like ...


0

LVM should only be used on a single system, if two nodes think they have exclusive access to an LVM volume corruption will occur. HA-LVM allows for multiple nodes to read from a shared volume but assumes only one node can/should write - this is analogous to how Windows MSDTC clustering worked/works (not that up to speed with in today sorry) where a single ...



Top 50 recent answers are included