Planning a Weka System Installation
Last updated
Last updated
The planning of a Weka system is essential prior to the actual installation process. It involves the planning of the following:
Total SSD net capacity and performance requirements
SSD resources
Memory resources
CPU resources
Network
A Weka system cluster runs on a group of hosts with local SSDs. To plan these hosts, the following information must be clarified and defined:
Capacity: Plan your net SSD capacity. Note that data management to object stores can be added after the installation. In the context of the planning stage, only the SSD capacity is required.
Redundancy Scheme: Define the optimal redundancy scheme required for the Weka system, as explained in .
Failure Domains: Determine whether failure domains are going to be used (this is optional) and if yes determine the number of failure domains and the potential number of hosts in each failure domain, as described in , and plan accordingly.
Hot Spare: Define the required hot spare count, as described in .
Once all this data is clarified, you can plan the SSD net storage capacity accordingly, as defined in the . You should also have the following information which will be used during the installation process:
Cluster size (number of hosts).
SSD capacity for each host, e.g., 12 hosts with a capacity of 6 TB each.
Planned protection scheme, e.g., 6+2.
Planned failure domains (optional).
Planned hot spare.
SSD resource planning involves how the defined capacity is going to be implemented for the SSDs. For each host, the following has to be determined:
The number of SSDs and capacity for each SSD (where the multiplication of the two should satisfy the required capacity per host).
The technology to be used (NVME, SAS, or SATA) and the specific SSD models, which have implications on SSD endurance and performance.
The total per host memory requirements is the sum of the following requirements:
Type
Per Host Memory
Fixed
2.3 GB
Core-based
2.3 GB for each Frontend core
3.3 GB for each Compute core
2.3 GB for each Drive/SSD core
SSD-based
HostSSDSize/10KB
Capacity requirement
See below
Reserved for Operating System
The maximum between 8 GB and 2% from the total RAM
Reserved for SMB/NFS services
8 GB
Reserved for RDMA
2 GB
The per-host capacity requirement is calculated with the following formula:
For Example: 12 hosts, 10 Weka system cores per host (6 for compute, 4 for SSDs), 100 TB SSD system with 512 TB total system capacity (with object store), average file size 64 KB.
The capacity requirement for the host will be calculated according to the following formula:
Consequently, the overall requirement per host is: 4.6 + 6 * 3.3 + 4*2.3 + (100TB/10KB)/12 + 6.3 +8 +8 = 56.73 GB
The Weka software on a client host requires 4 GB of additional memory.
The Weka system implements a Non-Uniform Memory Access (NUMA) aware CPU allocation strategy to maximize the overall performance of the system. The allocation of cores utilizes all NUMAs equally to balance memory usage from all NUMA nodes.
The following should be noted with regards to the CPU allocation strategy:
The code allocates CPU resources by assigning individual cores to tasks in a cgroup
Cores in a Weka cgroup won't be available to run any other user processes
On systems with Intel hyperthreading enabled, the corresponding sibling cores will be placed into a cgroup along with the physical ones.
The number of physical cores dedicated to the Weka software should be planned according to the following guidelines and limitations:
At least one physical core should be dedicated to the operating system; the rest can be allocated to the Weka software.
In general, it is recommended to allocate as many cores as possible to the Weka system.
No more than 19 physical cores can be assigned to Weka system processes.
Enough cores should be allocated to support performance targets.
In general, use 1 drive core per SSD for up to 6 SSDs and 1 drive core per 2 SSDs for more, with a ratio of 2 compute cores per SSD core.
For finer tuning, please contact the Weka Support Team.
Enough memory should be allocated to match core allocation, as discussed above.
The running of other applications on the same host (converged Weka system deployment) is supported. However, this is not covered in this documentation. For further information, contact the Weka Support Team.
On a client host, by default, the Weka software consumes a single physical core. If the client host is configured with hyper-threading, the Weka software will consume two logical cores.
If the client networking is defined as based on UDP, there is no allocation of core resources and the CPU resources are allocated to the Weka processes by the operating system as any other process.
Weka backend hosts can be connected to both InfiniBand or Ethernet networks. For each network technology used, all backends must be connected via this technology. If backends are connected both through Infiniband and Ethernet, the Weka system will favor the Infiniband links for traffic, unless there are connectivity issues with the Infiniband network. In that case, the system will use the Ethernet links (clients connecting to the system can connect either via Infiniband or Ethernet).
Client hosts can be configured with networking as above, which provides the highest performance and lowest latency, but requires compatible hardware and dedicated core resources. If compatible hardware is not available, or if allocating a physical core to the Weka system is problematic, the client networking can be configured to use the kernel UDP service. In such cases, performance is reduced, and latency increases.
On a dedicated host, all memory left after the reductions above is used for capacity. Otherwise, by default, weka host memory is set to 1.4 GB per compute-core, out of which 0.4 GB is used for the capacity requirement memory. If the default capacity requirement memory is not big enough to satisfy the total size of the filesystems, the least used metadata units will be paged to disk, as described in . If more RAM is desired for metadata, the must be performed in the install process. Having sufficient system memory is not enough.
