Racko

Frequently Asked Questions

Answers from our VPS, Dedicated Server, and GPU Cloud guides.

What is VPS hosting?

A VPS (virtual private server) is a virtual machine with its own dedicated share of CPU, RAM and storage on a physical server. You get root access and full control, without paying for a whole machine.

When should I choose VPS hosting over shared hosting?

Choose a VPS when your site or app needs guaranteed resources, custom software, root access or steady performance under traffic — things shared hosting cannot promise because every site on it competes for the same resources.

When should I choose a dedicated server?

When you need the full power of a physical machine: consistently high CPU or memory use, strict isolation or compliance needs, large databases, or workloads that outgrow a VPS.

What kind of workloads are suitable for a dedicated server?

High-traffic websites and e-commerce, large databases, ERP and business applications, game servers, video processing, virtualisation hosts and any workload that needs predictable, unshared performance.

What are the benefits of GPU cloud computing?

You get powerful GPUs on demand without buying hardware: faster AI training and inference, the ability to scale up for a project and down after it, and no upfront capital cost or maintenance.

Can I use dedicated servers for AI workloads?

Yes. CPU-based dedicated servers suit data preparation, classic machine learning and serving smaller models. For training or running large models, a GPU server is usually much faster.

How does GPU rental work?

You pick a GPU plan, pay for the period you need, and Racko sets up the server and emails your access details. You run your workloads on it like any other server, and renew or stop when you are done.

What is GPU cloud for AI training?

It is renting GPU servers to train machine learning models. GPUs process the large parallel calculations in training far faster than CPUs, cutting training time from weeks to days or hours.

How do I choose the right server for my business?

Start from your workload: a VPS for websites, apps and development; a dedicated server for heavy, steady workloads; a GPU server for AI and graphics. Then size CPU, RAM and storage to your peak usage. Our support team can help you decide.

What is GPU cloud for LLM inference?

It is running a trained large language model on GPU servers to answer requests. GPUs keep response times low and handle many requests at once, which CPUs struggle to do for large models.

Where can I find my invoices?

Sign in and open Billing in your dashboard. Each paid order there has its GST invoice ready to download.

Is GPU cloud more cost-effective than buying hardware?

For most teams, yes — especially for projects, experiments or variable workloads. You avoid the upfront cost, power, cooling and upgrades. Buying can make sense only when GPUs run at full use for years.

How do I contact support?

Visit our support page to chat with Racko Assist, message us on WhatsApp or raise a ticket. Support is available 24/7.

What industries can benefit from GPU cloud?

AI and software companies, research and education, healthcare imaging, finance and risk modelling, media and rendering, and any business building AI features into its products.

What is the difference between VPS and dedicated server?

A VPS is a virtual slice of a physical server with guaranteed resources; a dedicated server is the whole physical machine for you alone. Dedicated servers give more power and isolation; a VPS costs less and suits most websites and apps.

Do I need technical knowledge to use a VPS server?

Basic Linux or Windows server knowledge helps, since you manage the server yourself. Racko sets it up for you and includes monitoring and alerts, and our support team can help when you get stuck.

What factors affect GPU cloud pricing?

Mainly the GPU model and how many GPUs you need, plus the CPU, RAM and storage alongside them, and the billing period — longer terms usually cost less per month.

How do I compare price and value when choosing a server?

Look beyond the monthly price: compare guaranteed CPU, RAM and storage type, network speed, uptime commitment, support hours and what setup and monitoring are included. Cheaper plans often cost more in downtime.