Automation Glossary • Poll Group

What Is a Poll Group?

Merobix Engineering • • 4 min read

Reading a device one register at a time is slow and wasteful. A poll group is how SCADA bundles many points into a single request so each round trip does more work.

Back to Blog

Poll Group in one line: A poll group is a set of device points that SCADA reads together in one request rather than individually. By combining contiguous or nearby registers into a single message - for example one Modbus read covering registers 40001 to 40020 - the SCADA reduces the number of round trips on the network. Fewer, larger requests mean less protocol overhead, lower latency per point, and far better use of slow serial or cellular links.

Why grouping matters: the round-trip problem

Every poll is a request-response round trip. On a serial link at 9600 baud, or a cellular connection with hundreds of milliseconds of latency, the time cost is dominated by the round trip itself, not the size of the data. Reading twenty registers as twenty separate requests can take twenty times as long as reading them in one request, even though the payload is nearly identical.

A poll group solves this by describing a block of addresses the SCADA reads with a single command. The device returns all the values at once, and the SCADA driver distributes them to the individual tags internally. This is why grouping is one of the most effective ways to speed up a communications channel without changing hardware.

How poll groups are built

Most SCADA drivers build poll groups automatically. The driver looks at the addresses of the configured tags on a device, sorts them, and packs runs of contiguous or close addresses into blocks up to the protocol's maximum message size - for Modbus, up to 125 registers per read. Tags that fall in a gap are read in a separate block.

There is a trade-off. Grouping across a small gap reads a few unused registers but saves a whole round trip, which is usually worth it. Grouping across a large gap wastes bandwidth on registers nobody needs. Drivers expose a maximum-gap setting so the integrator can tune how aggressively addresses are merged. Sensible register layout in the device - keeping the points you scan often close together - makes grouping far more effective.

Poll groups in oil and gas field communications

Remote oil and gas sites almost always sit on bandwidth-constrained links: licensed radio, cellular, or satellite. Here poll grouping directly affects how many wells or tanks a single channel can cover. A poll cycle built from a few well-formed groups completes quickly and leaves headroom; a cycle of hundreds of single-register reads can stretch past the desired update interval and starve other sites on the same channel.

Poll groups are also tuned alongside scan rates. Fast-changing values like line pressure might be placed in a small, frequently polled group, while slow tank levels sit in a larger group polled every few minutes. Separating them keeps the fast group small and quick, so critical data is never delayed behind bulk reads of slow data.

When a flow computer or RTU is specified, engineers often lay out its register map deliberately so the SCADA's most-scanned points are contiguous. This is a quiet but real design decision: a good map lets the SCADA read a whole site in one or two efficient poll groups.

Frequently Asked Questions

Is a poll group the same as a scan class?

No. A scan class defines how often a set of tags is read (the timing). A poll group defines which points are physically bundled into one request message (the packing). A single scan class may be served by several poll groups, and tuning both together gives the best channel efficiency.

What limits how many points fit in one poll group?

The protocol's maximum message length. Modbus, for example, allows up to 125 holding registers per read. The driver also splits a group when addresses are too far apart, because bridging a large gap would waste bandwidth reading registers you do not use.

How do poll groups help on cellular links?

Cellular links have high latency per request, so fewer, larger requests are dramatically faster than many small ones. Good poll grouping cuts the number of round trips, which shortens the poll cycle and lets one channel serve more remote sites.

Sources and verification

This page references the protocol specifications published by the organizations below. Editions, product capabilities, and documentation change over time - confirm current requirements and specifications directly with the source.

Last reviewed: July 27, 2026. Merobix is not affiliated with, endorsed by, or sponsored by these organizations; their names are used only to identify the standards and products discussed.

From Definitions to a Live Dashboard

Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.

Request a Free Demo +1 (903) 307-7300
More in Automation Glossary
Heartbeat Signal  •  SCADA Gateway  •  Communications Watchdog  •  Unsolicited Response  •  Integrity Poll  •  Keepalive  •  All Automation Glossary →
Free SCADA operator training
Merobix University - 70 video lessons & 261 quiz questions, from first login to compliance reporting. No demo call required.
Start free →