Your data never transits our cloud
With the hybrid model, a lightweight agent inside your perimeter handles the data — while you manage scheduling and monitoring from our dashboard.
Glide your data home.
Elglide keeps your operational tables — and your views — flowing into Snowflake, continuously. Our hybrid model splits the work: an agent inside your network moves the data, while our cloud gives you the control plane — UI, scheduling, monitoring — and your data never transits Elglide's cloud.
- Free forever tier
- No credit card needed for registration
- Hybrid agent — data stays in your network
- Live in minutes, not weeks
Architecture
Control plane in the cloud. Data plane in your network.
This is the core of how Elglide is built, not an add-on mode: the control plane — scheduling, monitoring, self-tuning sync plans, the whole GUI — runs in our cloud. The data plane is a single agent you deploy inside your own network; it's the only thing that talks to your source databases and warehouse, and the only thing that ever touches a row. Elglide's cloud sees metadata and run status — never your data.
- Data plane stays in your network — the agent is the only thing that touches your rows
- Control plane in our cloud — full GUI for scheduling, monitoring, and tuning, nothing to self-host
- One agent per network, deployed in minutes
Don't need data residency? Elglide also runs fully in our cloud — no agent required.
Elglide cloud
Control plane
Schedule · monitor · audit — never a row
mutual TLS
Your firewall
outbound only · inbound: none
Yours
Your databasesSQL Server · PostgreSQL · MySQL · Oracle
read source
Elglide agentoutbound dialer
extract & load
Your Snowflakeyour own cloud account
Everything you need to move data safely
Six things Elglide handles so your team doesn't have to.
Flexible scheduling, per table
Near real-time, hourly, weekly, monthly, or strictly on-demand — set independently for every table or view.
Incremental, natively
CDC, change-tracking, watermarks, and RowHash diffing across SQL Server, PostgreSQL, MySQL, and Oracle.
Built for high throughput
Auto-scaling parallel extracts and automatic query-split keys handle large tables without hand-tuning.
Self-tuning pipelines
Recommendations on virtual primary keys, split keys, and audit thresholds — set automatically, adjustable anytime.
Predictable, transparent cost
Pay for rows actually moved — no credit card to start, and no surprise volume charges.
How it works
Three steps from source to warehouse — no infrastructure to stand up.
-
1
Connect your source
Choose SSH bastion, hybrid agent, or IP allowlist — whichever fits how your network is locked down.
-
2
Configure the sync
Pick a schedule per table, from near real-time to on-demand. Native CDC handles the incremental logic.
-
3
Land it, ready to use
Data lands in Snowflake as clean tables — your columns, nothing else — ready for BI and AI.
Security is the architecture, not a feature
With the hybrid model, your rows move straight from your databases to your Snowflake — only metadata and run status reach our cloud. Everything else follows from that.
- Hybrid agent handles data locally; you manage it from our dashboard
- SSH bastion and IP allowlist also available for stricter environments
- End-to-end data lineage and audit tracking on every sync
- Google and Microsoft sign-in included — SAML/OIDC with your own identity provider on request
Why Elglide
Elglide and competitors like Fivetran all move data from your operational databases into your warehouse. The difference is everything around that: what counts as a source, how the sync plan is chosen, how long your data lives outside your network, and what the bill looks like at the end of the month.
Views are first-class
Sync database views — full or incremental — exactly like tables. Fivetran doesn't support views at all.
Your data stays yours
In SaaS mode, staged data lives as compressed Parquet for seconds — purged, guaranteed. In hybrid mode, zero customer data ever touches the Elglide cloud.
Up to 95% lower cost
Transparent pricing — you pay for rows synced, shown live in-app, not an opaque Monthly Active Rows meter. At 5M changed rows a month that is $120 on Elglide against about $2,500 at Fivetran's published Standard rate — see the worked example.
| Feature | Elglide | Fivetran |
|---|---|---|
| Sync engine | ||
| Database views as sources | ✓First-class — full and incremental sync for views, powered by RowHash | ✕Not supported |
| Sync mechanism selection | ✓Detected automatically at runtime — Full, Watermark, RowHash, or log-based CDC — and refined over time | Fixed at connector setup; changing modes often means a full re-sync |
| Watermark incremental sync | ✓Yes — for both tables and views | ✕Not offered |
| Delete detection | ✓Every sync mode catches deletes — watermark and RowHash included, no log access required | ✕CDC-enabled connectors onlyAirbyte and Hevo Data have the same gap — deletes are only caught in CDC/log-based mode, not standard incremental sync. Stitch can't detect hard deletes at all without log-based replication. |
| Delta sync without touching the source | ✓RowHash — no log access, no schema changes, and it works on views | ✕No equivalent — depends on database log access |
| Very large tables (1B+ rows) | ✓Parallel split loads handle any scale, billions of rows per table | Initial syncs can stall or fail past roughly a billion rows per table |
| Destination & data trust | ||
| Clean destination schema | ✓Your columns, nothing else | ✕Reserved system columns added to every table, such as _FIVETRAN_DELETED |
| Customer data retention in vendor cloud | ✓SaaS mode: purged within a minute. Hybrid mode: zero customer data held | Buffered in the vendor's cloud for hours to days |
| Operations & governance | ||
| Scheduling granularity | ✓Database-level by default, customizable per table or per view | One schedule per connector |
| Audit & SOX compliance | ✓Built-in audits before, during, and after every pipeline run | Post-hoc sync logs |
| No-code experience | ✓Drag-and-drop pipeline canvas — no technical background required | Form-based configuration aimed at data engineers |
| Cost | ||
| Pricing model | ✓Transparent rows-synced pricing — every billed row visible live in-app (worked example) | Monthly Active Rows (MAR) — bill scales with data churn, hard to forecast; volume discounts unpublished |
Comparison based on publicly available competitor documentation and pricing as of September 2026; capabilities may have changed since. Fivetran, Airbyte, Hevo Data, and Stitch are trademarks of their respective owners and are not affiliated with Elglide.
One simple price. Pay for what you sync.
No tiers to compare — just credits for the rows you move, on every connectivity option, including the hybrid agent.
Pay as you go, after your free rows
$30per 1M rows synced
Buy credits as you need them — a balance you load never expires.
Included for every account
- Hybrid agent, SSH bastion & IP allowlist
- Unlimited users
- Real-time to on-demand scheduling
- Native CDC & incremental syncing
- Self-tuning sync plans
Always free
Full syncs
The initial full load of a table or view — however many rows it contains — never counts toward your usage. Not on day one, not ever.
Metered
Incremental syncs
Only what changes afterward — new, updated, or deleted rows — counts toward your total, at the rate above.
What the bill looks like — a worked example
One database connection. "Rows changed per month" is what Elglide meters after the free full load, and roughly what Fivetran counts as Monthly Active Rows.
| Rows changed / month | Elglide | Fivetran Standard |
|---|---|---|
| 500,000 | $0 — inside the 1M free rows | $0 — inside the 500K-MAR Free plan |
| 850,000 | $0 — inside the 1M free rows | $423.78 — Fivetran's own published example (847,574 MAR) |
| 5,000,000 | $120 — 4M billable rows × $30 per 1M | About $500 for the first 1M MAR alone at the same published rate; the remaining 4M at volume rates Fivetran does not publish |
Sources: Fivetran's pricing page, September 2026 — Free plan 500,000 MAR; Standard-plan examples of $17.23 for 34,479 MAR and $423.78 for 847,574 MAR (about $0.50 per 1,000 MAR); a $5 monthly minimum per connection; volume discounts above 1M MAR per connection are not published. "Up to 95% lower" is the saving at 5M rows if Fivetran's published rate applied throughout; at their discounted volume rates the saving is smaller. Your own figure depends on how many rows change each month — the usage meter in the app shows it live.
Load once, use whenever: credit never expires, and you can load up to $5,000 at a time so a busy month never interrupts a sync. We charge your card only when you choose to load — no card on file, no automatic charges.
Need single sign-on with your own identity provider, an uptime SLA, priority support, or a dedicated deployment?
Talk to us about add-ons →Register and deploy the hybrid agent in 20 minutes
Get the full SaaS platform running with your data still inside your network.