HPC Rental — Enterprise High Performance Computing Infrastructure
Enterprise HPC rental services in Türkiye. On-premises or remote GPU/CPU cluster rental, SLURM deployment and managed operations. KVKK compliant, infrastructure located in Türkiye.
There are two ways to meet enterprise computing requirements: buy or rent. The Mevasis HPC rental model gives your organisation scalable, managed high performance computing capacity without a capital-intensive hardware purchase.
Why Rent HPC?
High performance computing infrastructure is a decision with serious technical and financial weight. Hardware cost, floor space, cooling, power draw and specialist staff — together they frequently push past the budget that was planned for them.
The rental model hands all of that over to Mevasis:
Scalability: Capacity grows with your workload. Extra GPU nodes come online for project peaks and go back when the project closes.
Capital preservation: OPEX instead of CAPEX. Depreciation schedules, balance sheet weight and the risk of technology ageing all disappear.
Managed operations: Hardware maintenance, software updates, security patching and monitoring are run by the Mevasis team.
KVKK compliance: With infrastructure located in Türkiye your data stays in-country. For projects that process personal data, legal exposure is minimised.
Rental Models
On-Premises HPC Rental
Mevasis hardware is installed in your own facility. Network integration, user management and workload optimisation are handled by the Mevasis team.
Best suited to: projects where data residency is critical — healthcare, defence, finance and the public sector.
Remote (Dedicated) Rental
A private cluster allocated to you in Mevasis data centres, reached securely over VPN or a dedicated network link.
Best suited to: organisations that need large-scale compute capacity without holding physical space.
CPU Hour Rental
A pay-as-you-go model billed per core-hour consumed. Suited to irregular workloads and research projects.
GPU Hour Rental
Hourly access to NVIDIA H100, A100 or L40S GPUs for AI model training, simulation and visual rendering.
Technical Configuration Options
Mevasis proposes a configuration from the following hardware pool according to your requirement:
| Component | Options |
|---|---|
| CPU node | AMD EPYC 9004 (Genoa), Intel Xeon Sapphire Rapids |
| GPU node | NVIDIA H100 SXM5, A100 80GB, L40S |
| Memory | 256 GB – 1.5 TB per node |
| Storage | BeeGFS / Lustre parallel file system |
| Network | InfiniBand HDR200/NDR400 or 25/100GbE |
| Scheduler | SLURM (default), PBS, SGE |
What Every Package Includes
- Hardware installation and commissioning
- SLURM or your preferred scheduler, configured
- User account and access management
- Monitoring (Prometheus + Grafana)
- Software environment (Lmod, MPI libraries, core scientific software)
- SLA-backed support (24/7 for critical, business hours for standard)
- Monthly capacity report
How It Works
- Requirements analysis: the Mevasis team reviews your workload profile, user count and compute requirements
- Quote: technical specification and pricing within 48 hours
- Contract: service level agreement and rental terms agreed
- Deployment: hardware supply, installation and commissioning — typically 2–6 weeks
- Go live: user training and handover
For more information and a tailored quote, use the contact form or call +90 212 483 70 33.
Frequently Asked Questions
Contracts run from a one-month minimum up to multi-year terms. Project-based short-term rentals come with additional flexibility.
Renting requires no capital investment, is accounted for as operating expenditure and scales with demand. When you buy, responsibility for the entire hardware lifecycle stays with you.
Yes. With Mevasis infrastructure located in Türkiye your data never leaves the country, which keeps you fully aligned with KVKK requirements.
Core count, memory capacity, GPU type, storage requirement and contract length are the main factors. Fill in the quote form for detailed pricing.
Related Solutions
Let's Build Future Together.
