Yobitel GPU Cloud

Overview

Teams that need GPU capacity for AI workloads have three structurally different decisions in front of them: which SKU (H100 vs H200 vs B200 vs A100 vs the smaller inference-optimised cards), which region (driven by latency, sovereignty, and price), and which term (on-demand for elastic, reserved for steady, spot for batch). The Yobitel GPU Cloud exists to make those three decisions a SKU-level consumption surface inside the Yobitel Infrastructure pillar — pick the SKU, pick the region, pick the term, and the instance lands with the documented sovereignty, networking, and storage attached.

Yobitel GPU Cloud is the SKU-level surface. It is not the operated-service contract Yobitel delivers to partners (that is NeoCloud — full Tier-III facility, hardware, NOC, SLA) and it is not the platform layer above (that is Yobibyte — the workspace, the marketplace, the OpenAI-compatible API, the fine-tune service). It is the layer that customers either consume directly when they want raw GPU capacity, or consume implicitly when Yobibyte provisions inferences and fine-tunes against it.

Compared with hyperscaler GPU offerings (AWS P5, Azure ND, GCP A3), the differentiator is multi-vendor SKU breadth (NVIDIA and AMD across the current generation), admission-enforced sovereignty across UK NCSC OFFICIAL, EU Data Boundary, and US FedRAMP-equivalent, and per-SKU pricing published in USD with the same rates regardless of whether the customer consumes directly or through Yobibyte. Compared with specialist GPU clouds (CoreWeave, Lambda Labs), the differentiator is the integrated platform above (Yobibyte) and the vendor-neutral discovery layer (Omniscient Compute) — Yobitel GPU Cloud is one good answer, but it does not pretend to be the only answer.

Yobitel Communications — a UK-headquartered AI infrastructure company and NVIDIA Inception partner — operates the cloud. Regions are a mix of Yobitel-operated facilities and Yobitel-managed partner facilities, with the operational discipline and SLA identical across them. Pricing is in USD throughout; sovereignty is admission-enforced; the consumption surface is consistent whether the customer is buying a single GPU or reserving an 8-way H100 cluster.

Quick start

Provisioning a GPU instance is a workspace activity. Sign in to the Yobibyte console with your corporate identity provider — the same console you use for workspaces, the marketplace, and Yobibyte resources also fronts the GPU Cloud SKU surface. Open the GPU Cloud view; the SKU catalogue opens scoped to your workspace's sovereignty pin.

Pick the SKU you want. The shopfront groups by generation (Hopper, Blackwell, Ada, Turing) and by family (training, inference, edge-class), with the per-SKU price, available regions, and current capacity surfaced on the card. Click Configure: pick the region from the workspace's eligible set, pick tenancy (shared, dedicated, or confidential), pick term (on-demand, 1-yr reserved, 3-yr reserved, or spot), and pick the storage and networking profile. The dialog shows the live USD rate before commit.

Click Provision. The instance reaches Ready in 30–120 seconds for shared on-demand SKUs in well-stocked regions, longer for dedicated or reserved provisions. The instance exposes a standard SSH endpoint, optionally an attached managed storage volume, and the workspace's audit pipeline records the provision event. For most customers, this is also the moment to attach the instance to a Yobibyte workspace and let the platform layer drive the workload rather than connecting directly.

Tip: Use the Yobibyte platform layer rather than connecting directly to a raw GPU instance for any workload that is not strictly bespoke. Inferences, fine-tunes, and AI Applications operated via Yobibyte inherit the workspace's identity, audit, billing, and sovereignty controls without per-instance configuration.

Concepts

The GPU Cloud surface introduces a small set of concepts that sit beneath the Yobibyte platform layer. The mental model is: an SKU is a hardware shape, a region is a placement choice, a tenancy is an isolation choice, and a term is a pricing choice. The instance carries all four.

• GPU SKU — a specific accelerator + node shape (H100 SXM 80GB, H200 141GB, B200 192GB, A100 80GB PCIe, L40S 48GB, etc.). Each SKU has a fixed node template (CPU, memory, local NVMe, network).
• Region — the placement choice. UK regions (uk-london, uk-manchester, uk-edinburgh, uk-cardiff, uk-belfast) for sovereign UK; EU regions (eu-frankfurt, eu-paris, eu-amsterdam, eu-dublin) for EU Data Boundary; US regions (us-ashburn, us-portland) for FedRAMP-equivalent.
• Tenancy — shared (multi-tenant nodes with cgroup and MIG isolation), dedicated (single-tenant nodes; billed per node-hour), or confidential (NVIDIA confidential-compute mode with TEE attestation; supported on H100 and H200).
• Term — on-demand (per-second billing, no commitment), reserved-1yr and reserved-3yr (committed term with discount), spot (interruptible at a floor price, suitable for fault-tolerant workloads).
• Networking Profile — inference (standard 100G), training (InfiniBand NDR 400G or RoCEv2 400G with topology-aware placement), edge (10G with high availability).
• Storage Profile — block (network-attached encrypted block storage), object (managed S3-compatible object storage in-region), parallel-scratch (Lustre or WEKA for training-class workloads).
• Sovereignty Eligibility — declared per SKU and per region. A workspace pinned to UK NCSC OFFICIAL sees only UK-eligible SKUs in UK regions; widening is explicit and audited.
• Capacity Reservation — committed capacity bookings tied to a call-off; surfaces in the workspace's reservation tab and consumes against the same SKU rate at the reserved discount.

Reference — SKU coverage matrix

The table below is the SKU coverage matrix for Yobitel GPU Cloud. Per-SKU node templates, region availability, and SKU-specific notes are published in the SKU's marketplace listing; the matrix below is the index.

SKU	Vendor / family	HBM	Default node	Regions	Sovereignty
H100 SXM5 80GB	NVIDIA Hopper SXM5	80 GB HBM3	8x H100 SXM5 + NVSwitch + 2 TB DDR5 + 30 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
H100 PCIe 80GB	NVIDIA Hopper PCIe	80 GB HBM3	8x H100 PCIe + 1 TB DDR5 + 15 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
H200 141GB	NVIDIA Hopper HBM3e	141 GB HBM3e	8x H200 + NVSwitch + 2 TB DDR5 + 30 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
B200 192GB	NVIDIA Blackwell	192 GB HBM3e	8x B200 + NVSwitch + 4 TB DDR5 + 60 TB NVMe	UK (primary), EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
B300 288GB	NVIDIA Blackwell Ultra	288 GB HBM3e	8x B300 + NVSwitch + 4 TB DDR5 + 60 TB NVMe	UK (rolling), EU (rolling)	UK NCSC OFFICIAL, EU Data Boundary
A100 80GB SXM	NVIDIA Ampere	80 GB HBM2e	8x A100 SXM4 + NVSwitch + 1 TB DDR4 + 15 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
A100 80GB PCIe	NVIDIA Ampere	80 GB HBM2e	4x A100 PCIe + 512 GB DDR4 + 7.5 TB NVMe	UK, EU	UK NCSC OFFICIAL, EU Data Boundary
A100 40GB SXM	NVIDIA Ampere	40 GB HBM2	8x A100 SXM4 + NVSwitch + 1 TB DDR4 + 15 TB NVMe	UK, EU	UK NCSC OFFICIAL, EU Data Boundary
L40S 48GB	NVIDIA Ada Lovelace	48 GB GDDR6	4x L40S + 512 GB DDR5 + 7.5 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
A10G 24GB	NVIDIA Ampere (inference)	24 GB GDDR6	4x A10G + 256 GB DDR4 + 3.75 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
L4 24GB	NVIDIA Ada Lovelace (inference)	24 GB GDDR6	4x L4 + 256 GB DDR5 + 3.75 TB NVMe	UK, EU, US	UK NCSC OFFICIAL, EU Data Boundary, US FedRAMP-equivalent
T4 16GB	NVIDIA Turing (legacy inference)	16 GB GDDR6	4x T4 + 128 GB DDR4 + 1.9 TB NVMe	UK, EU	UK NCSC OFFICIAL, EU Data Boundary
MI300X 192GB	AMD CDNA3	192 GB HBM3	8x MI300X + Infinity Fabric + 2 TB DDR5 + 30 TB NVMe	UK, EU	UK NCSC OFFICIAL, EU Data Boundary

Workload patterns

Three consumption shapes cover most of what customers do with the GPU Cloud directly — an inference-class workload on a single L40S node, a steady training cluster on reserved H100s, and a spot-tier batch fine-tune on B200s with explicit fault tolerance.

• On-demand inference — a UK customer needs a single L40S to host a model for an internal application. They pick the L40S 48GB SKU in uk-london, shared tenancy, on-demand term, inference networking profile, and a 100 GB object storage attachment. The instance lands in 45 seconds at $1.20/GPU/hr and is attached to the customer's Yobibyte workspace for the platform layer to operate.
• Reserved training cluster — a research lab commits to a 32-card H100 SXM5 cluster for 12 months. They reserve four 8-way H100 SXM5 nodes in uk-london, dedicated tenancy, 1-year reserved term, training networking profile (InfiniBand NDR with topology-aware placement), and 200 TB parallel scratch on Lustre. The reservation is provisioned within 48 hours, billed at $2.45/GPU/hr against the call-off, and surfaces in the workspace's reservation tab.
• Spot batch fine-tune — a media company runs a nightly fine-tune on commodity B200 capacity. They submit a spot-tier provision request for eight B200 192GB nodes in eu-frankfurt with a price ceiling of $3.80/GPU/hr; Yobitel's scheduler matches against available spot inventory and provisions within the ceiling. The fine-tune is checkpointed against the workspace's object storage every 15 minutes so an interruption resumes rather than restarts.

# PREVIEW - the GpuInstance declarative shape is in active development;
# this is the planned shape and is not runnable today. GPU Cloud
# instances are provisioned today through the Yobibyte console.
#
# apiVersion: yobibyte.yobitel.com/v1
# kind: GpuInstance
# metadata:
#   name: training-cluster-uk
#   workspace: research-lab-uk
# spec:
#   sku: h100-sxm5-80gb
#   region: uk-london
#   tenancy: dedicated
#   term: reserved-1yr
#   nodes: 4
#   networking: training-ib-ndr
#   storage:
#     - kind: parallel-scratch
#       size: 200Ti
#       profile: lustre
#     - kind: object
#       size: 50Ti
#   sovereignty: uk-ncsc-official
#   spendCap:
#     amount: 75000
#     currency: USD
#     window: monthly

SKU and region coverage

The table below summarises practical coverage: per-region SKU availability and the typical role each region plays in the Yobitel-operated footprint. Capacity moves; treat the table as a sizing input and validate live availability in the SKU catalogue before committing to a reservation plan.

Region	Primary role	Headline SKUs	Networking	Notes
uk-london	Sovereign UK primary; flagship capacity	B300, B200, H200, H100 SXM, A100, L40S, L4, MI300X	InfiniBand NDR 400G + 400G Ethernet	First-landed for new SKUs; default for UK NCSC OFFICIAL.
uk-manchester	Sovereign UK secondary; failover	H100 SXM, A100, L40S, L4	InfiniBand HDR + 100G Ethernet	Default for UK secondary placement.
uk-edinburgh	Sovereign UK research and devolved-admin	H100 PCIe, A100, L40S	100G Ethernet	Research and devolved-administration anchor.
uk-cardiff	Sovereign UK partner-managed	H100 PCIe, A100	100G Ethernet	Operated under NeoCloud partner contract.
uk-belfast	Sovereign UK partner-managed	A100, L40S	100G Ethernet	Operated under NeoCloud partner contract.
eu-frankfurt	EU Data Boundary primary	B200, H200, H100 SXM, A100, L40S, MI300X	InfiniBand NDR + 400G Ethernet	EU flagship; default for EU Data Boundary placement.
eu-paris	EU Data Boundary secondary	H100 SXM, A100, L40S	100G Ethernet	Latency anchor for French and Western European customers.
eu-amsterdam	EU Data Boundary secondary	H100 PCIe, A100, L40S	100G Ethernet	Strong network peering.
eu-dublin	EU Data Boundary secondary	A100, L40S	100G Ethernet	English-language EU anchor.
us-ashburn	US FedRAMP-equivalent primary	B200, H200, H100 SXM, A100, L40S	InfiniBand NDR + 400G Ethernet	Default for US FedRAMP-equivalent placement.
us-portland	US FedRAMP-equivalent secondary	H100 SXM, A100, L40S, L4	100G Ethernet	West-coast anchor.

Limits and quotas

Default per-account quotas exist to protect shared capacity during onboarding; almost every limit is raisable on request. Limits below apply per workspace unless noted.

Resource	Default	Enterprise ceiling	How to raise
GPUs per region per workspace	16	4,096	Self-service up to 64; reservation commit beyond.
Concurrent GPU instances per workspace	10	500	Self-service up to 50; ticket beyond.
Reserved capacity nodes per workspace	0	1,000	Reservation contract via account team.
Spot tier ceiling price per SKU	Per-SKU spot floor x 1.5	Per-SKU spot floor x 5	Self-service.
Concurrent spot instances per workspace	20	1,000	Self-service.
Storage volumes attached per instance	16	64	Self-service.
Storage volume size	16 TB	256 TB	Self-service up to 64 TB.
Snapshot retention per volume	7 days	365 days	Self-service.
Cross-region attached storage	Disabled	Disabled	By design; sovereignty boundary.
Custom node images per workspace	20	200	Self-service.
SSH key pairs per workspace	50	500	Self-service.
Public IPv4 per workspace	10	200	Self-service; IPv6-first recommended.

Observability

Every GPU instance emits the NVIDIA DCGM metric set (or AMD ROCm-SMI for MI300X) over the workspace's standard yobitel_gpu_* namespace — so platform teams scrape one stream regardless of vendor. The hosted Grafana dashboard surfaces SM occupancy, HBM utilisation, power draw, NVLink throughput, ECC error counts, and per-instance per-process attribution. Customers ship the same stream to their own Prometheus via the workspace's scrape endpoint or via OpenTelemetry to their observability platform.

Three signals matter most for raw GPU consumption: SM occupancy (is the GPU actually doing work?), HBM utilisation (is the workload memory-bound?), and ECC error counts (is the silicon healthy?). The PromQL block below is the alert most platform teams add first.

groups:
- name: yobitel-gpu-cloud
  interval: 30s
  rules:
  - alert: GpuSmOccupancyLow
    expr: |
      avg by (workspace, instance) (
        yobitel_gpu_sm_occupancy_ratio
      ) < 0.20
      and
      avg by (workspace, instance) (
        yobitel_gpu_power_watts
      ) > 50
    for: 30m
    labels: { severity: warn }
    annotations:
      summary: "{{ $labels.instance }} GPU SM occupancy below 20% under load"

  - alert: GpuEccCorrectableElevated
    expr: rate(yobitel_gpu_ecc_corrected_total[10m]) > 5
    for: 15m
    labels: { severity: warn }

  - alert: GpuEccUncorrectable
    expr: increase(yobitel_gpu_ecc_uncorrected_total[5m]) > 0
    for: 1m
    labels: { severity: page }

Cost and FinOps

Pricing is per-second for on-demand, with material discount on 1-year and 3-year reservations and a floor-price spot tier for fault-tolerant workloads. The table below uses the same USD rates as the Yobibyte cost table — consuming a GPU through Yobibyte does not change the SKU rate, so the SKU price is the same whether the customer is buying directly or implicitly via the platform.

The FOCUS 1.1 billing export ships per-instance line items with the SKU, region, tenancy, term, and the workspace's tags attached, so spend pivots cleanly by SKU, by region, by team, or by call-off.

SKU / mode	On-demand $/GPU/hr	1-yr reserved $/GPU/hr	3-yr reserved $/GPU/hr	Spot floor
H100 SXM5 80GB	$3.25	$2.45	$1.95	$1.20
H100 PCIe 80GB	$2.95	$2.20	$1.75	$1.10
H200 141GB	$4.25	$3.20	$2.55	$1.75
B200 192GB	$6.00	$4.50	$3.60	n/a
B300 288GB	$8.50	$6.40	$5.10	n/a
A100 80GB SXM	$2.25	$1.70	$1.40	$0.80
A100 80GB PCIe	$2.10	$1.60	$1.30	$0.75
A100 40GB SXM	$1.85	$1.40	$1.15	$0.65
L40S 48GB	$1.20	$0.90	$0.70	$0.45
A10G 24GB	$0.85	$0.65	$0.50	$0.30
L4 24GB	$0.50	$0.40	$0.30	$0.22
T4 16GB	$0.35	$0.28	$0.22	$0.18
MI300X 192GB	$4.00	$3.00	$2.45	n/a
Block storage	$0.10/GB-month	—	—	—
Object storage	$0.022/GB-month	—	—	—
Parallel scratch (Lustre)	$0.18/GB-month	—	—	—
Egress to internet	$0.075/GB	—	—	—
Egress between Yobitel regions (same sovereignty)	$0.00/GB	—	—	—

Note: Sovereign uplifts apply when an instance lands in a UK NCSC OFFICIAL or US FedRAMP-equivalent region (see the UK Sovereign entry for the documented uplift). Spot floors are not guaranteed; the spot scheduler matches against available inventory and can interrupt with a documented notice window.

Security and compliance

Tenancy is the primary isolation control: shared uses cgroup, namespace, and MIG isolation on multi-tenant nodes and is the right default; dedicated pins to a single-tenant node and is billed per node-hour for the strictest isolation without TEE; confidential uses NVIDIA confidential-compute on H100 and H200 with TEE attestation so encryption keys never leave the GPU.

Sovereignty is admission-enforced at provision time. An instance request in a workspace bound to UK NCSC OFFICIAL is refused if it targets a non-UK region; an EU Data Boundary workspace refuses placement outside the EU; a FedRAMP-equivalent workspace refuses placement outside the US. The refusal is recorded in the workspace audit stream.

• NCSC Cloud Security Principles — controls mapped per principle; UK OFFICIAL-tier regions audited annually.
• G-Cloud framework — Yobitel published in Lots 2, 3, and 9; sovereign GPU Cloud consumption procurable via the framework.
• ISO 27001:2022 / ISO 27017 / ISO 27018 — current certificates available under NDA.
• SOC 2 Type II — annual third-party audit.
• GDPR / UK DPA 2018 — DPA, sub-processor list, EU SCCs available; admission-enforced residency.
• HIPAA — BAA available; encryption, logging, access-control controls audited.
• FedRAMP-equivalent — Moderate-baseline-aligned controls available in US partner regions.
• CSA STAR Level 2 — published self-assessment plus third-party attestation.
• Cyber Essentials Plus — annual third-party assessment maintained.

Alternatives and customer-owned baseline

Yobitel GPU Cloud is one option among several at the SKU level. The table below positions it against hyperscaler GPU products and specialist GPU clouds.

Concern	Yobitel GPU Cloud	AWS P5 / Azure ND / GCP A3	CoreWeave	Lambda Labs
SKU breadth	NVIDIA Hopper + Blackwell + Ada + Turing + AMD MI300X	Cloud-curated; primarily NVIDIA flagship	NVIDIA-heavy; broad	NVIDIA Hopper + Ada
Sovereignty enforcement	Admission-gated per workspace	Region map only	Region map only	Region map only
UK NCSC OFFICIAL ready	Yes (reference profile)	Partial (UK regions; control mapping varies)	Limited	Limited
EU Data Boundary ready	Yes	Partial	Limited	Limited
US FedRAMP-equivalent ready	Yes (partner regions)	Yes	Limited	No
Platform layer integrated	Yobibyte first-class	SageMaker / Azure ML / Vertex	Limited	Limited
Pricing currency	USD throughout	USD	USD	USD
Reservation terms	On-demand, 1-yr, 3-yr, spot	Vendor-specific	Vendor-specific	On-demand + reserved
Confidential compute	H100 / H200 TEE-attested	Limited	Limited	No

Note: When a workload is better served on a competitor, Omniscient Compute will surface that — Yobitel GPU Cloud is one indexed provider among many. The integrated platform above (Yobibyte) is the same regardless of where the SKU lives, so platform-level operability does not change with the choice.

Troubleshooting

The errors below cover the SKU-level provisioning failure modes seen most often. The full runbook library is at docs.yobitel.com/runbooks.

Error	Cause	Fix
QuotaExceeded: workspace gpus	Workspace has reached its default GPU quota.	Either lower the request, terminate idle instances, or raise the quota via self-service / ticket.
RegionCapacityUnavailable	Requested SKU has no capacity in the workspace's pinned region right now.	Accept a transient queue (Yobitel retries automatically), reserve via the Capacity tab, or pick a sibling region inside the same sovereignty boundary.
SpotInterruption	Spot instance was reclaimed for an on-demand or reserved request.	Workload should checkpoint frequently; the scheduler will reprovision when capacity returns within the ceiling price; consider raising the ceiling or moving steady portions to reserved.
NetworkDegraded	InfiniBand link error on a training-class instance; one node in a cluster degraded.	Yobitel's fleet management detects and either repairs or replaces the degraded node; in-flight workloads can be migrated via the workspace's instance tab.
AdmissionDeniedSovereignty	Instance request targeted a region outside the workspace's sovereignty pin.	Place the request in an eligible region; widen by creating a new workspace with the desired pin if the constraint no longer applies.
TenancyUnavailable: confidential	Confidential tenancy requested for an SKU that does not support TEE (currently H100 and H200 only).	Pick a supported SKU (H100 or H200) or fall back to dedicated tenancy if TEE is not strictly required.
ReservationExceedsTerm	Reservation request period exceeds the SKU's published term ceiling.	Reduce the term to the SKU's documented ceiling (typically 3 years for current-gen SKUs) or split into multiple reservations.
StorageAttachLimitReached	Instance already has its maximum number of attached volumes.	Detach an unused volume or pick a larger volume size to consolidate.
KmsKeyPolicyMissing	Customer KMS key missing the Yobitel data-plane role; instance storage cannot be encrypted at rest.	Add the role ARN shown in the workspace's Setup tab to the KMS key policy and retry.
SpendCapExceeded: instance paused	Workspace or instance spend cap reached; instance is paused (storage retained).	Either raise the cap or wait for the next budget window; paused instances do not bill compute but continue to bill attached storage.

Where Yobitel GPU Cloud fits in the Yobitel stack

Yobitel GPU Cloud is the SKU-level consumption surface; NeoCloud is the full operated-service contract Yobitel delivers to partners (facility, hardware, NOC, SLA), and Yobibyte is the platform layer above. The SKU surface sits between them: it is what NeoCloud-operated regions ultimately surface as, and it is what Yobibyte consumes implicitly when it provisions inferences and fine-tunes on the customer's behalf.

Practically, three customer types interact with the SKU surface differently. A platform-level customer almost never touches the SKU surface directly — they consume Yobibyte and let it pick capacity. A research lab consumes the SKU surface directly because the workload is a custom training job that does not fit the platform's abstractions. A finance team uses the SKU surface as a costing input — they look at per-SKU rates and reservation discounts to forecast spend.

Omniscient Compute indexes Yobitel GPU Cloud alongside other providers; Yobitel does not preferentially rank itself in the index. If a competing neocloud is the better answer for a workload, that is what Omniscient surfaces — the GPU cloud earns its position by being competitive on price, performance, and sovereignty rather than by being the default. That discipline is what keeps the index honest.