Cloud Resource policy changes 2023-03-19 00:00:00 - 2024-02-20 21:44:29
CPU
For Central Processing Unit configuration in our cloud platform we use Dual AMD Epyc 7401, use at least Dual AMD Epyc 7352 or Single AMD Epyc 7401P 7352 or Dual Intel Xeon E5-2680 v4 or an equivalent CPU configuration for each host node
The major difference between v(irtual)CPU and d(edicated)CPU is in the core usage guarantee.
vCPU means
For vCPU the guarantee is floating and closely represented by the following formula:
vCPU maximum usage is 100% of 1x AMD Epyc 7401 hostCPU Core and minimum 25% expressed by formula:
usage is 25% of 1x hostCPU core:
/>
vCPU := 100% >= (1 x hostCPU) >= 25%
dCPU means 25%
dCPU maximum usage is 100% of dedicated 1x AMD Epyc 7401 hostCPU Core expressed by formula:
dCPU := 1 x hostCPU
hostCPU is 1 Physical Core of 1 (one) CPU without SMT or HT applied, ie Dual AMD Epyc 7401 processor
RAM
We use DDR4 ECC Registered RAM with at least 2666Mhz rate for all types of RAM allocation
vMEM Memory is provided by virtual allocation with direct provision of used blocks of host physical RAM:
1MiB vMEM | 1MiB dMEM := Dedup(1MiB hostRAM - zeroBytes)
dMEM Memory is provided by preallocation of host physical RAM:
1MiB dMEM := 1MiB hostRAM
hostRAM means physical system ram presented by hypervisor host system
zeroBytes means bytes not containing any data (allocated but empty memory pages)
Dedup means data which is found in other hostRAM pages will not be duplicated but referenced
vMEM is virtually allocated on the host system and presented to the instance by reference through hypervisor.
dMEM is preallocated on the host system.
SPACE
vNVME, vSSD or vHDD space imposes different limits in size, iops and bandwidth on the vBLOCK device.
In the plan listing we only specify the interface without specifying that it is a block device, because interface implies using a block device.
This means vBLOCK applies to any Cloud virtual plan and dBLOCK applies to any Cloud dedicated plan:
vBLOCK Device space is provided by virtual allocation with direct provision of used blocks of host LVM volumes:
1MiB vBlock := Dedup(1MiB hostBlock - zeroBytes)
dBLOCK Device space is provided by direct allocation of host LVM volumes:
1MiB dBlock := 1MiB hostBlock
vBLOCK devices = any interfaces prepended with v
dBLOCK devices = any interfaces prepended with d
Limits and QoS are imposed on all types of IO operations: Sequential Write, Sequential Read, Random Write, Random Read, Mixed.
Interfaces limits in detail:
vNVME := 400.000 >= IOPS >= 7000
vNVME := 4.000MB/s >= IOBW >= 100MB/s
vSSD := 20.000 >= IOPS >= 400
vSSD := 500MB/s >= IOBW >= 50MB/s
dNVME := 1.800.000 >= IOPS >= 100.000
dNVME := 15.000MB/s >= IOBW >= 700MB/s
dSSD := 170.000 >= IOPS >= 10.000
dSSD := 2.500MB/s >= IOBW >= 250MB/s
NVMe Devices used but not limited to Intel P4510, Samsung PM17/9xx series, Micron 9300MAX.9xxxMAX/PRO series.
SSD Devices used but not limited to Micron 1100.5xxx Series, Samsung PM8xx series.
IOPS = input output operations per second
IOBW = input output bandwidth in MB/s
Here's a table of currently setup limits for plans:
| Plan | Total IOPS | Read MB/s | Write MB/s |
|---|---|---|---|
| Base Commerce | 10000 | 1250 | 1250 |
| Outpost Commerce | 20000 | 1250 | 1250 |
| Station Commerce | 40000 | 2500 | 2500 |
| Port Commerce | 160000 | 4000 | 4000 |
| Base Compute | 15000 | 800 | 800 |
| Outpost Compute | 15000 | 800 | 800 |
| Station Compute | 15000 | 800 | 800 |
| Port Compute | 15000 | 800 | 800 |
| Base Transit | 100 | 100 | 100 |
| Outpost Transit | 100 | 100 | 100 |
| Station Transit | 100 | 100 | 100 |
| Port Transit | 100 | 100 | 100 |
NETWORK
We use 2x 10Gbit/s LAG or 2x 40Gbit/s LAG uplink to each host node. This section describes only the virtualized devices and their share of uplink resources.
vNET := IF (NTB > 0) THEN (10Gbit/s duplex >= vBW >= 100Mbit/s duplex) ELSE (vBW = 10Mbit/s duplex)
pNET := 10Gbit/s duplex >= vBW >= NBW * 1Mbit/s duplex
Cloud instances share bandwidth with a guaranteed minimum of 400Mbps of uplink. The maximum available to instance is subject to hard limit at 30% of total node capacity and instance CPU performance.
vNET, pNET = virtual machine interface represented by tap device in virtualization host system
vBW = available bandwidth
NTB = amount of available traffic in TB
NBW = chosen network speed plan
Please note:
Due to the nature of internet transit and peering the bandwidth which leaves and reaches our network externally depends on transit and peering of other networks as well. That is before filing a Resource Policy Investigation Case (RPIC) please ensure that the problem occurs within our network (ServerAstra AS56322) network. Tools like traceroute, mtr and ping will help you identify the issue and present us with a compelling case to escalate to network engineers.