# IOIT — Digital Infrastructure Engineering Partner (full reference for LLMs)

This is the long-form companion to /llms.txt. It contains the full canonical text of every primary service IOIT delivers, so language models can quote concrete claims with confidence and link back to the right URLs.

## Company facts

- Legal name: IOIT (ТОО IOIT)
- Founded: 2019
- Headquarters: Almaty, Kazakhstan, проспект Достык 210, 050051
- Coordinates: 43.2389°N, 76.8897°E
- Phone: +7-747-499-77-89
- Email: mc@ioit.co
- Hours: Mon–Fri 09:00–18:00 Asia/Almaty (UTC+5)
- Languages: Russian (primary), Kazakh, English
- Vendor partnerships: Juniper Networks (Elite Partner), Fortinet, Zextras (Carbonio), Timex; plus ZTE, Marvel, ARMO, HPE, Dell, Lenovo, Supermicro, Cisco, Huawei, NVIDIA, NetApp, Pure Storage, Nutanix, VMware per project
- Track record: 100+ projects delivered, 10 in-house engineers, 8 years on market, 0 customer claims, 99.97% SLA uptime
- Service area: Kazakhstan (Almaty, Astana, Shymkent, Atyrau, Ust-Kamenogorsk and other cities) and CIS

## Service 1 — Network & Connectivity (https://ioit.co/services/network)

We design and deploy enterprise- and carrier-grade networks: campus LANs, data-centre fabrics (EVPN-VXLAN, Spine-Leaf), branch SD-WAN, Wi-Fi 6/6E, and secure access (ZTNA, SASE). Each network is built to measurable business metrics — latency, throughput, availability.

Deliverables:
- Campus & DC networks: design and deployment of EVPN-VXLAN fabrics, Spine-Leaf topologies, VRF segmentation, automation via Ansible/NetBox.
- SD-WAN & branch networks: branch connectivity, multi-ISP, application-aware routing, encryption (Fortinet Secure SD-WAN, Juniper Session Smart Router).
- Wi-Fi 6/6E: RF planning, controllers, seamless roaming, guest portal, high user density and IoT.
- ZTNA · SASE · NAC: Zero Trust Network Access, identity-based access control, secure remote access for hybrid teams.

Stack: Juniper Mist AI, Fortinet Security Fabric, EVPN-VXLAN, Spine-Leaf, SD-WAN, SASE, Wi-Fi 6/6E, BGP/OSPF/IS-IS, Ansible, NetBox, ZTNA, NAC, IPv6, TIA-568, ISO/IEC 11801.

## Service 2 — Data Centre & Infrastructure (https://ioit.co/services/datacenter)

We design new data centres and modernise existing sites: from concept and BIM model to choosing reliability tier (Uptime Tier II–IV) and optimising PUE. We help build hybrid infrastructure and migrate workloads between on-prem, colocation, and cloud without prolonged downtime.

Deliverables:
- Data-centre design: concept, BIM model, PUE calculation, power/cooling/SCS/security subsystems. TIA-942, Uptime Tier, ASHRAE TC 9.9.
- Site modernisation: audit, targeted subsystem upgrades, hot/cold aisle conversion, density and redundancy increase.
- Workload migration: on-prem ↔ colocation ↔ public cloud, step-by-step plan, zero/minimal downtime, RTO/RPO control.
- Colocation & hybrid: placement in partner Tier III DCs in Kazakhstan, connectivity, unified operations plan on-prem + colocation + cloud.

Stack: TIA-942, Uptime Tier II–IV, ASHRAE TC 9.9, PUE/DCIM, Schneider/Vertiv/Eaton, APC InfraStruXure, BIM/Revit, VMware/Proxmox/Nutanix, AWS/Azure/GCP.

## Service 3 — Server & Storage Solutions (https://ioit.co/services/server-storage)

We pick and deploy server infrastructure and storage systems sized to your workload profile: classic three-tier (compute · storage · network), hyperconverged (HCI), and unified storage with block, file, and object access. We design backup and disaster-recovery policy with RPO/RTO aligned to business goals.

Deliverables:
- Server platform: 1U/2U, blade, HCI; CPU choice (x86, ARM), memory, NICs by workload profile. HPE, Dell EMC, Lenovo, Supermicro.
- Storage & SDS: all-flash and NVMe arrays, scale-out storage, software-defined storage (Ceph, vSAN). Unified block/file/object.
- Backup & DR: 3-2-1 strategy, immutable repositories, DR site, restore drills (Veeam, Commvault, Bacula).
- Virtualisation & HCI: VMware vSphere, Proxmox VE, Nutanix, Microsoft Hyper-V; live migration between hypervisors.

Stack: HPE/Dell/Lenovo/Supermicro, NetApp/Pure Storage, Nutanix/vSAN, Ceph/MinIO, VMware/Proxmox, Veeam/Commvault, NVMe/NVMe-oF, FC/iSCSI, S3/NFS/SMB.

## Service 4 — Automation & DevOps (https://ioit.co/services/automation)

We turn infrastructure into code, releases into a pipeline, and operations into an observable process. We roll out Infrastructure-as-Code, GitOps, container orchestration, and an observability stack so teams ship faster and incidents get caught before they hit the business.

Deliverables:
- Infrastructure-as-Code: Terraform/Pulumi for cloud and on-prem; Ansible for configuration. Reproducible environments, Git review.
- CI/CD & GitOps: GitLab CI / GitHub Actions / Jenkins; ArgoCD/Flux deployment; automated tests; canary releases.
- Kubernetes & containers: cluster design and operations, on-prem and managed; service mesh, secrets, security policy.
- Observability & SRE: Prometheus, Grafana, Loki, Tempo, ELK; SLO/SLI; error budgets; on-call.

Stack: Terraform, Pulumi, Ansible, Salt, Kubernetes, Helm, Docker, containerd, ArgoCD, Flux, GitLab CI, GitHub Actions, Prometheus, Grafana, Loki, Tempo, ELK, Istio, Linkerd, Vault, OPA.

## Service 5 — AI & Data Intelligence (https://ioit.co/services/ai)

We build infrastructure and platforms for AI/ML: GPU clusters, MLOps pipelines, production inference of LLMs and classical models, enterprise RAG, and data platforms. We help teams move from pilot notebooks to production with measurable quality and cost control.

Deliverables:
- AI infrastructure: GPU nodes (NVIDIA H100/A100/L40S), distributed-training network fabric, GPU sharing across teams via Kubernetes.
- MLOps: data and model versioning, experiment tracking, CI/CD for models, drift and quality monitoring (MLflow, Kubeflow, DVC).
- LLM, RAG & inference: enterprise RAG over private data, private self-hosted inference, vector DB, latency/cost-per-token optimisation.
- Data platform: lakehouse storage, ETL/ELT orchestration, BI, self-service analytics (ClickHouse, Postgres, Spark, Airflow, dbt, Superset).

Stack: NVIDIA CUDA, Triton Inference Server, vLLM, TGI, PyTorch, TensorFlow, Hugging Face, MLflow, Kubeflow, DVC, LangChain, LlamaIndex, pgvector, Qdrant, Milvus, ClickHouse, Spark, Airflow, dbt, Superset.

## Service 6 — Managed Services & Support (https://ioit.co/services/managed)

We take infrastructure operations off your plate and run them turnkey: 24×7 NOC, monitoring, incident and change management, regular preventive work, on-site engineers. Service level is fixed in the SLA — with clear response and recovery times.

Deliverables:
- 24×7 NOC & monitoring: on-shift NOC, single monitoring system, proactive alerts, runbooks.
- Incident management: logging, prioritisation, escalation per ITIL/ISO 20000; post-mortems on majors.
- SLA & proactive operations: tier matched to criticality, regular maintenance, restore drills, transparent KPI reporting.
- On-site engineers: field engineers across Kazakhstan, spares stock, 4-hour windows for critical systems.

Stack: ITIL, ISO/IEC 20000, SLA/OLA, 24×7 NOC, Zabbix, PRTG, Grafana, Prometheus, Jira Service Management, PagerDuty, Opsgenie.

## Engagement process (six stages)

1. Brief — 1–2 days. Objective, context, constraints, budget.
2. Spec & quote — 3–7 days. Technical spec, commercial proposal, work plan.
3. Design — 2–6 weeks. Working and as-built documentation, approvals.
4. Supply — 3–12 weeks. Equipment sourcing, delivery, warehousing.
5. Implementation & integration — 2–8 weeks. Deployment, configuration, automation, integration tests.
6. Hand-off — 1–2 weeks. Acceptance testing, training, SLA activation.

## FAQ

Q: Do you only work in Almaty?
A: We cover all of Kazakhstan — on-site work in Astana, Shymkent, Atyrau, Ust-Kamenogorsk and other cities. SLA 24×7 applies across the entire country.

Q: How long does a typical project take?
A: From brief to hand-off — 3–6 months for a medium facility (2–5k m² office, 1–3 network nodes). Large malls and industrial sites — 8–12 months.

Q: Do you work with public-sector clients?
A: Yes. We participate in tenders via Goszakupki, FIK, and closed RFPs. Documentation follows TEO / PSD / RD format per Kazakhstan GOSO requirements.

Q: Can we engage just one of the services — say, only the network or only AI infrastructure?
A: Yes. Each of the six directions (network, data centre, server & storage, automation & DevOps, AI & data, managed services) is a standalone service. Take one direction or combine them.

Q: What post-delivery support do you provide?
A: Up to 2-year warranty on all work. After that — an extended SLA per contract: 8×5 or 24×7, with on-site dispatch and remote monitoring.

Q: Which vendors do you work with directly?
A: Juniper Networks (Elite Partner), Zextras (Carbonio), Timex, plus ZTE, Marvel, ARMO, HPE, Cisco, Dell — per project requirements.

## Canonical URLs

- Home: https://ioit.co/
- Services hub: https://ioit.co/services
- Network & Connectivity: https://ioit.co/services/network
- Data Centre & Infrastructure: https://ioit.co/services/datacenter
- Server & Storage Solutions: https://ioit.co/services/server-storage
- Automation & DevOps: https://ioit.co/services/automation
- AI & Data Intelligence: https://ioit.co/services/ai
- Managed Services & Support: https://ioit.co/services/managed
- Juniper catalogue: https://ioit.co/products/juniper
- Fortinet catalogue: https://ioit.co/products/fortinet
- Projects: https://ioit.co/projects
- About team: https://ioit.co/about-team
- MEGA Silk Way case: https://ioit.co/cases/mega-silkway
- Sitemap (XML): https://ioit.co/sitemap.xml