AWS EC2 Instances›Memory Optimized›x8i›x8i.48xlarge

Compared with Rackspace Spot

AWSRackspace Spot
Machinex8i.48xlarge22 × mh.vs1.2xlarge
vCPU192352
Memory3072 GiB2816 GiB
Local diskNoneNone
On-demand / hr$21.008$9.482
Spot / hr$4.306$0.462
Prices in US East, N. Virginia

Compute

vCPUs192
Memory3072 GiB
Mem / vCPU16.00 GiB
ProcessorXeon Scalable
Clock speed3.9 GHz
Architecturex86_64
GPU0
GPU Wattage0 W
GPU ArchitectureNone
Video Memory0 GiB
FPGA0
ffmpeg FPSN/A
CoreMark / secN/A

Storage

TypeEBS only
Bandwidth60,000 Mbps
Throughput7500 MB/s
IOPS240,000
Base Bandwidth60,000 Mbps
Base Throughput7500 MB/s
Base IOPS240,000
Devices0
NVMe DriveNo
SSDNo
Swap PartitionNo

Networking

Performance75 Gigabit
Max ENIs24
IPv6Yes
GPUNone
Placement GroupSupported
TrunkingYes
Branch Interface98
IPs / ENI64
Baseline Bandwidth75 Gbps

NUMA Architecture

Uses NUMANo
Node CountN/A
Max DistanceN/A
Cores / NodeN/A
Threads / NodeN/A
Memory / NodeN/A
L3 Cache / NodeN/A
L3 Cache SharedN/A

AWS Metadata

Instance typex8i.48xlarge
Full nameX8I 48xlarge
FamilyMemory optimized
Variantx8i
GenerationCurrent
Bare MetalNo
EMR supportNo
DedicatedSupported
Introduced2026-01-15
Family sizes

x8i.48xlarge

AWS's x8i.48xlarge is a memory-optimized instance with 192 vCPU and 3072 GiB. Rackspace Spot's largest server is 16 vCPU and 128 GiB, so matching this instance would take 22 × mh.vs1.2xlarge. That is arithmetically true and operationally absurd.

The Rackspace Spot column below prices that whole fleet, so read it as a scale reference rather than a drop-in replacement: at this size the useful question is what a shard of the workload costs, which is $0.021/hr per server.

Server comparison caveats
No single-node equivalent

The AWS x8i.48xlarge is 192 vCPU and 3,072 GiB. The largest comparable Rackspace Spot server is mh.vs1.2xlarge at 16 vCPU and 128 GiB, so matching it would take 22 of them. That maths works out, but 22 nodes is a redesign rather than a migration, because a single block of 3,072 GiB does not survive being split across a network.

If you chose x8i.48xlarge because the work genuinely needs 192 cores and 3,072 GiB together (an in-memory database, a large analytics engine, a single big solver), Rackspace Spot has no equivalent. What does fit is everything around it: the application servers, batch workers and CI that talk to that machine, which is usually the larger share of the bill.

Pricing by region

1 hr

Regions are Rackspace Spot locations.

RegionAWS on-demandAWS Spot1-yr Reserved3-yr ReservedRackspace on-demandRackspace SpotOn-demand price savingsSpot price savings
US East, N. Virginia$21.008$4.306not offerednot offered$9.482$0.462
$11.526 (55%)
$3.844 (89%)
US Central, Chicagono AWS regionn/an/an/a$10.208$0.462n/an/a
US Central, Dallasno AWS regionn/an/an/a$10.186$2.882n/an/a
Hong Kongnot sold heren/an/an/a$10.868$0.462n/an/a
Sydney, Australia$30.462$3.046not offerednot offered$10.802$4.202
$19.660 (65%)
AWS cheaper
London, UKnot sold heren/an/an/a$9.240$0.638n/an/a
US West, San Josenot sold heren/an/an/a$9.680$0.044n/an/a
US Central, Dallas 2no AWS regionn/an/an/a$9.680$5.324n/an/a

Prices updated 17 Nov 2026.

When to use it

Good for
  • ●Large JVM heaps and application servers that hold state
  • ●OLAP and business intelligence query engines
  • ●Real-time analytics over hot datasets
  • ●Workloads that fit inside 3072 GiB of RAM
  • ●Parallel work that can occupy 192 vCPU
  • ●Network-heavy services (AWS rates this size at 75 Gigabit)
Avoid if
  • ●The working set needs more than 3072 GiB of RAM
  • ●You want a single-machine Rackspace Spot equivalent. At this size there is not one.
  • ●CPU-bound work. You are paying for memory this workload will not use.
  • ●More than 16.00 GiB of RAM per vCPU is needed, since that ratio is fixed for the whole family
  • ●The work is single-threaded. Most of these 192 vCPU would sit idle.
  • ●Local NVMe or instance-store disk is a hard requirement
x8i.48xlarge is $15,336/mo on AWS on-demand. A single mh.vs1.2xlarge is $315/mo on Rackspace Spot on-demand. At this size the per-server price is the honest comparison.
  • Rackspace Spot sells spare capacity at market prices. It is the cheapest compute on the internet.
  • Prices are set by supply and demand. You pay what the market decides.
  • Compute is delivered as fully managed Kubernetes. All infrastructure is taken care of.
  • Egress costs $0.01/GB after the free allowance. AWS charges $0.09/GB.
Build your Cloudspace

Frequently asked questions

How much does x8i.48xlarge cost?

The AWS x8i.48xlarge costs $21.01/hr on-demand in US East (N. Virginia), about $15,336/mo running 24/7, or $4.31/hr at the AWS Spot price. On Rackspace Spot, 22 × mh.vs1.2xlarge is $9.48/hr on-demand, or $0.462/hr at the Spot price. Rackspace Spot also sells in Chicago and Dallas, where AWS operates no region at all, so those rows have no AWS price to compare against.

How much is x8i.48xlarge per month?

At 730 hours a month, x8i.48xlarge in US East (N. Virginia) is $15,336/mo on AWS on-demand, or $3,143/mo at the AWS Spot price. The same capacity on Rackspace Spot (22 × mh.vs1.2xlarge) is $6,922/mo on-demand, or $337/mo at the Spot price. Rackspace Spot prices move with demand, so the Spot figure is a typical rate rather than a fixed one.

Can I replace x8i.48xlarge with Rackspace Spot?

Not as a single machine. Rackspace Spot's largest VM shape is 16 vCPU and 128 GiB, so it would take 22 × mh.vs1.2xlarge to reach 192 vCPU and 3072 GiB, and past roughly twenty nodes that stops being a credible substitution. Where the workload does shard (batch, web, queue consumers), a single mh.vs1.2xlarge at $0.021/hr is the number to compare against; the table's Rackspace Spot column prices the full 22-node fleet.

What is a cheaper alternative to x8i.48xlarge?

Rackspace Spot 22 × mh.vs1.2xlarge (352 vCPU, 2816 GiB) is $6,922/mo on-demand or $337/mo at the Spot price, against $15,336/mo on-demand or $3,143/mo at the Spot price for x8i.48xlarge on AWS in US East (N. Virginia). The trade is real: Rackspace Spot capacity is priced by demand and can be reclaimed when demand rises, and workloads have to be containerized.