What It Means To Count SAP HANA Memory In Data Center Capacity

TL;DR

Prime Big Deal Days · Oct 6–7Offer from Amazon

Get school and study supplies delivered free — and shop member deals

• Fast, free delivery on millions of items
• Access to Prime Big Deal Days deals on October 6–7
• Prime Video, Amazon Music and more included

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

Rymvard released version 0.23 on October 4, 2026, adding SAP HANA memory accounting, placement proposals and replication takeover checks to its data center capacity ledger. The company tested the early-access software against a simulated landscape; it has not reported a customer deployment or real-world outcomes for this release.

Rymvard released version 0.23 on October 4, adding SAP HANA memory accounting and replication takeover checks to its data center capacity ledger, as detailed in the original analysis. The early-access software is aimed at SAP hosting providers and operations teams that need to track whether hosts and backup sites can support HANA systems; the company’s reported tests used a simulated estate, not a customer deployment.

The release divides each HANA host’s physical memory among the operating-system reserve, memory used by HANA systems, allocated but unused capacity, memory reserved for replicas from other sites, and remaining free memory. Rymvard sets the default operating-system reserve at 10% of the first 64 GiB and 3% of the remaining memory, unless a host has a different reserve configured. System usage is based on its 30-day peak. If allocations exceed physical memory, the ledger reports overbooking rather than showing negative free capacity.

For SAP HANA system replication, the software checks whether a secondary host could accommodate the primary system’s allocation after a takeover. It reports whether the workload fits, fits after named test or development systems are stopped, does not fit (with the shortfall in GiB), or cannot be assessed. Production systems belonging to other customers are not treated as stoppable. A replication status older than one hour or a memory reading older than 24 hours leads to an unknown result, rather than a passing assessment. Scale-out systems are evaluated as a whole, using the weakest host pair.

Version 0.23 also proposes a host and replication-site partner for a new HANA system and estimates quarterly memory runway at each site, weighting planned go-lives by their probability. A new SAP Operations workspace highlights free HANA memory, production systems without replication, takeovers that would fail, and hosts above 90% utilization. Rymvard says HANA instances appear in its compute view so teams can include SAP memory in customer showback and lifecycle planning.

Making HANA Failover Capacity Visible

The development addresses a planning problem for providers whose immediate constraint is memory capacity, not the overall power or space available in a data center. A site can have room in its broader infrastructure while a particular HANA host—or its replication partner—lacks the memory needed for a new allocation or takeover. Separating overbooking, free memory and replica reservations may help operations teams identify those limits before placing a system or relying on failover capacity.

The checks are planning signals, not proof that a takeover will succeed under every operating condition. Their usefulness depends on accurate and current source data, configuration and workload behavior. Rymvard’s stale-data rule errs on the side of caution by marking old readings unknown, but that can leave teams needing to verify capacity through other operational processes. The placement and runway views may support site comparisons; the material provided does not establish savings, improved reliability or customer outcomes.

How the SAP Memory Ledger Works

Rymvard says version 0.23 reads information from SAP through read-only interfaces. Its collector connects to SAP Host Agent over HTTPS and is limited to four read operations: ListInstances, ListDatabases, GetDatabaseStatus and GetComputerSystem. A separate monitoring connection uses TLS and a fixed set of six SELECT queries, with a 60-second limit per system and no more than eight systems queried at once. The company says the software does not write to SAP.

The release also lists read-only information sources for landscape details: SAP Focused Run, ServiceNow, SAP Solution Manager through its System Landscape Directory, and SAP LaMa. These are presented as supported sources, not as tools that must all be used together. Rymvard reports testing the release against a simulated landscape containing 200 HANA hosts, 161 systems and 39 replicated pairs. In that scenario, it identified one production system without replication, a takeover short by 512 GiB, a host overbooked by 128 GiB and a replication in an error state. The company says displayed screens use an illustrative estate and do not represent a customer environment.

“Nothing is ever written to SAP.”

— Rymvard

Customer Evidence Is Still Pending

Rymvard describes version 0.23 as early access and has not identified a customer deployment or provided customer outcome data for this release. The simulated test results do not establish how the software performs across live customer landscapes, unusual SAP configurations or changing workloads. No independent evaluation is included in the supplied release material.

The company has not announced a public price list or a date for broader availability; it says early-access terms are arranged directly with partners. The release also does not detail how every possible SAP configuration is handled, how integrations are validated in each environment, or how readings update beyond the stated age thresholds. Those points remain open for prospective users to assess.

Early-Access Evaluation Ahead

The next stated step is evaluation by early-access partners, with pricing arranged directly with Rymvard. The company has not announced a broader release date, named a customer using version 0.23, or published customer results. Evidence from live deployments would be needed to show how the memory ledger and takeover assessments perform against real workloads and operational changes.

For now, organizations considering the release can compare its documented read-only access limits and freshness rules with their SAP environments, then determine whether the available data and integrations support their planning needs. Rymvard has not said when further product or evaluation details will be published.

Key Questions

What did Rymvard add in version 0.23?

It added SAP HANA memory accounting, host and replication-site placement proposals, takeover-capacity checks, and quarterly memory-runway estimates to its data center capacity ledger.

Has Rymvard tested the software with a customer?

The reported version 0.23 test used a simulated landscape of 200 HANA hosts, 161 systems and 39 replicated pairs. The company has not identified a customer deployment or supplied customer outcomes for this release.

How does the software handle old capacity data?

Rymvard says the result becomes unknown if replication status is more than one hour old or a memory figure is more than 24 hours old. It does not count stale information as evidence that a takeover will fit.

Does Rymvard write data to SAP?

Rymvard says its SAP connections are read-only and states, “Nothing is ever written to SAP.” Its release description specifies read operations for the SAP Host Agent and a separate monitoring connection using SELECT queries.

When will the software be generally available?

No broader availability date has been announced. The company says the immediate next step is evaluation with early-access partners, with pricing agreed directly.

Primary source: Rymvard · via ThorstenMeyerAI.com