# BIG-IP DNS (GTM) GSLB simulator

> Model two-tier global server load balancing: a wide IP picks a pool, then the pool picks a virtual server. See how N DNS requests resolve across pools and members.

- Tool: https://ronutz.com/en/tools/bigip-dns-gslb-simulator
- Family: Networking

---

## What it does

Models BIG-IP DNS (formerly GTM) global server load balancing, which is two-tier: a wide IP first selects a pool, then the pool selects a virtual server. Configure your pools - each with a ratio, a region tag, an up/down state, and a member-selection method - and their members, then set the wide-IP pool method, a client region, and a request count N. The tool computes how the next N DNS name-resolution requests resolve: first the per-pool distribution, then the per-member distribution within each pool.

## What it simulates, and what it does not

Deterministic at both tiers: Round Robin (circular and even), Ratio (weighted, largest-remainder), Global Availability (all requests to the first available resource in list order, the rest as backups), and Topology (the resource whose region matches the client wins with the highest score; ties inside the winning score round-robin). A down pool or member receives nothing, and the traffic redistributes among those that are up.

The dynamic methods - Quality of Service, Completion Rate, Round-Trip Time, Fewest Hops, Kilobytes/Second, Packet Rate, Virtual Server Score, Least Connections, CPU - depend on live performance metrics that the big3d agents collect from each data center. Those cannot be reproduced offline, so the tool says so plainly rather than inventing a distribution. This is the same honesty rule the LTM simulator follows for its dynamic methods.

## Why two tiers matter

The most common GSLB confusion is treating pool selection and member selection as one decision. They are separate, and each can use a different method: a wide IP might pick pools by Topology (send European clients to the EU pool) while each pool picks a virtual server by Ratio (weight the bigger site heavier). The tool makes that separation visible - you see which pool answered, then which virtual server inside it resolved - so the interaction between the two tiers is concrete rather than abstract.

## Standards and references

- [F5 — Global Server Load Balancing (BIG-IP DNS)](https://techdocs.f5.com/en-us/bigip-15-0-0/big-ip-dns-load-balancing/global-server-load-balancing.html) - tiered GSLB: pool then virtual server; static vs dynamic methods
- [F5 — About load balancing and resource availability](https://techdocs.f5.com/kb/en-us/products/big-ip-dns/manuals/product/big-ip-dns-load-balancing-14-0-0/02.html) - wide-IP pool ordering; Round Robin, Ratio, Global Availability, Topology, QoS
- [F5 — Using Topology Load Balancing to Distribute DNS Requests](https://techdocs.f5.com/en-us/bigip-15-0-0/big-ip-dns-load-balancing/using-topology-load-balancing-to-distribute-dns-requests-to-specific-resources.html) - topology-record scoring; ties round-robin within the winning score
- [F5 Cloud Docs — tmsh gtm pool a (alternate/fallback modes)](https://clouddocs.f5.com/cli/tmsh-reference/v15/modules/gtm/gtm_pool_a.html) - the full method set and the three-tier preferred/alternate/fallback model

## Related reading

- [GSLB in Two Tiers: The Wide IP Picks a Pool, the Pool Picks a Server](https://ronutz.com/en/learn/gslb-two-tier-pool-then-member.md): The single most common BIG-IP DNS misunderstanding is treating global load balancing as one decision. It is two: a wide IP first selects a pool, then that pool selects a virtual server - and each tier can use a different method. Walking the two-tier decision with a worked example.