Dedicated Solana node — bare metal, single tenant, from $2,900/mo.

Our dedicated AMD EPYC 7543p Solana node is $2,900/mo flat, and right now your first month is 50% off — just $1,450 to start, then $2,900/mo. With yearly billing the effective price drops to $2,436/mo — a 16% discount that saves you $5,568 per year. There are no credit-metering charges, no overage bills, and the Yellowstone gRPC plugin is included for free. For comparison, Helius dedicated nodes start at $2,900/mo, and RPC Fast lists a 1.15 TB node from $2,700/mo in EU/NA, with a 3-month minimum (published prices, checked 2026-10-19).

Yes — new dedicated nodes get 50% off the first month on monthly billing. You pay $1,450 to start instead of $2,900, then it renews at the standard $2,900/mo. There's no long-term commitment on monthly billing, so it's a low-risk way to benchmark a single-tenant node against your current provider before committing. If you'd rather lock in the lowest ongoing rate, yearly billing is $2,436/mo effective (16% off).

Each node is a single-tenant bare-metal AMD EPYC 7543p: 32 CPU cores, 1024 GB of DDR4 ECC RAM, 2x 3.84TB NVMe (ledger + accounts), and a 10 Gbps uplink, unmetered. Nothing is shared with another tenant — CPU, RAM, NVMe, and bandwidth are 100% yours.

It is a dedicated server — single-tenant bare metal, an AMD EPYC 7543p — with the Solana client, the ledger snapshot and the Yellowstone plugin installed and kept running by us. If you would rather rent an empty server and run Solana yourself, the self-hosting costs are broken down line by line on our dedicated node cost page.

Yes. The node serves only your traffic, on your own hostname and API key: a standard JSON-RPC endpoint plus, if you keep the plugin, a Yellowstone gRPC endpoint. No other customer's requests touch the machine.

Both. Agave is the default Anza-maintained validator client and is recommended for the vast majority of workloads. Jito is the MEV-focused fork — pick it only if you specifically need bundle simulation (simulateBundle) or want to participate in MEV-flavoured RPC patterns. You can switch client at any time on request.

We currently provision in Frankfurt (Germany) and Singapore. Frankfurt is the natural choice for European and US-East traders; Singapore is the lowest-latency option for APAC. If you need a region we don't list yet, keep an eye on our Discord and Twitter for new hardware arrivals.

Within 3 working days of payment. Every node is set up by hand: we provision the server, sync the ledger snapshot, install your chosen client and the Yellowstone plugin, then send the endpoint, API key and onboarding steps by email and Telegram.

Yes — the Yellowstone gRPC Geyser plugin is bundled at no extra cost. You get a private gRPC endpoint streaming slots, blocks, transactions, and account update notifications, in addition to a full standard JSON-RPC endpoint. If you only need RPC features you can opt out of the plugin during checkout.

Self-hosting a comparable rig runs $3,500–5,000/mo all-in once you add hardware lease, 10 Gbps bandwidth, hot-standby for HA, snapshot management, and DevOps on-call. Our $2,900/mo (or $2,436/mo on yearly) is a turnkey replacement: same single-tenant hardware, with the operational headaches handled. See /solana-validator-geyser-setup for the full self-host breakdown if you want to compare.

Pay in USDC on Solana. You deposit USDC into your Subglow account balance from the dashboard, then complete the purchase — exactly the same flow as our shared gRPC + RPC plans. No card, no wire, no invoice round-trip.

Yes — yearly billing is 16% off, which works out to $5,568 saved per year on the EPYC 7543p (effective $2,436/mo vs $2,900/mo monthly). You can switch from monthly to yearly at any renewal — the discount is locked in for the full prepaid period.

Most traders and bot builders don't, and we would rather say so than sell you one. A flat-rate shared gRPC + RPC plan handles sniping, copy-trading, and analytics for the vast majority of workloads. You need a dedicated node when you have a strict single-millisecond latency budget, compliance rules that forbid shared infrastructure, or sustained request volume that a shared endpoint rate-limits. If none of those apply, shared is the cheaper and faster-to-start choice.

Not automatically, and this is the most common misconception about them. A well-colocated shared endpoint in the right region routinely beats a dedicated node in the wrong one, because network distance dominates. Dedicated hardware wins on latency only when you remove noisy-neighbour variance and colocate near your bot. For most teams the biggest gains come from picking a provider close to their execution infrastructure and from server-side filtering — not from dedicating hardware.

Our nodes are bare metal, so there is no virtualisation layer or neighbouring tenant adding jitter. Sub-millisecond round trips depend on distance more than hardware, though: your bot has to run close to the node, ideally in the same datacenter. We provision in Frankfurt and Singapore, so if your bot already runs in one of those cities, ask us in chat which datacenter your node will be in before you buy. If you need your execution engine in the same rack as the validator, that is the case where self-hosting wins.

On shared infrastructure your requests share CPU, memory bandwidth, and network with other tenants. Under heavy cluster load a spike in someone else's traffic can add jitter to your tail latency. A reputable provider isolates and over-provisions to minimise this, but a single-tenant node removes the variance entirely — every cycle is yours.

Yes, and it is what we recommend. Because Subglow serves standard Yellowstone gRPC on both shared and dedicated, moving up is an endpoint and key change — your subscription and filter logic are unchanged. Start shared, measure your real latency and volume, and move to dedicated only when the data says you need it.