NplusT provides 3D NAND characterization services led by engineers with senior-level experience in NAND technology and storage applications. Using its NanoCycler platform, the team helps customers investigate device behavior, assess reliability, and interpret results in the context of engineering and storage requirements.
As 3D NAND devices become more complex, collecting measurements is only part of characterization. Engineering teams also need to understand what the data indicate and how that behavior depends on:
NplusT’s engineers work with customers to define characterization plans, conduct measurements, analyze results, and identify functional and data-integrity margins.
A useful characterization program begins with the decisions a customer needs to make in light of the application requirements. A device team may be studying changes associated with cycling. A storage team may be investigating retention, read recovery, or error-rate trends. A reliability group may need repeatable measurements to compare samples or explore failure mechanisms.
NplusT helps scope the work around those objectives. The goal is to produce measurements that are traceable to the conditions under which they were collected and meaningful for the engineering question.
The service goes beyond reporting measurements. NplusT’s engineers help customers interpret trends, understand the limits of the available data, and determine whether follow-up measurements could clarify the results. This can help teams across device development, reliability engineering, controller design, and storage qualification make better-informed decisions.
At the device level, NAND reliability is reflected in cell behavior and errors observed during read operations. Storage systems manage those effects through technologies and policies that may include read-retry procedures, error-correcting code (ECC), data management, and firmware.
RBER measurements provide information about raw errors under defined conditions. Interpreted alongside cycling, retention, and read-retry results, they can help teams assess how device behavior evolves and identify questions for further analysis. RBER alone does not determine product-level outcomes, which also depend on the device, controller, ECC implementation, firmware, workload, and operating conditions.
Read-retry behavior can provide evidence about changes in cell-state distributions and read margin. Studying it alongside error-rate and retention data helps teams investigate NAND response to defined stress conditions and consider the relevance to storage design.
Endurance and retention are also related. Program/erase cycling can change device characteristics, while retention studies examine how stored data evolves over time and under specified conditions. Characterization helps teams observe these effects systematically and use the evidence to guide reliability assessment and further device or system-level evaluation.
NanoCycler provides the platform for agreed test sequences and measurements. It supports structured flash memory testing and data collection; NplusT’s engineers bring the expertise to shape the investigation and interpret its results.
NplusT collaborates with semiconductor companies, storage solution developers, and research organizations to align test plans with their objectives. This may include agreeing on sample handling, conditions, measurement sequences, and data outputs. Customers contribute device context and application requirements; NplusT provides characterization capability and engineering analysis.
By combining NanoCycler-enabled testing with senior-level experience in 3D NAND and storage applications, NplusT helps customers build an evidence-based understanding of endurance, retention, read-retry, RBER, and related reliability questions.
In addition to characterizing memories for typical storage applications and workloads, there are several customer applications that require “non-conventional” memory use, and therefore need dedicated characterization.
Use of NAND for caching with low-cost data persistency, required a unique high endurance/low retention design point. A first exploration of the retention/endurance trade-off increased the rated endurance of the 3D TLC by 3x. A further in-depth endurance characterization of the 3D TLC RBER distribution by layers identified the “offending” layers that failed (exceeded ECC correction capabilities) first.
By excluding these layers the Endurance was increased by 10x at the target retention time at the cost of less than 1% of NAND capacity, with a correspondingly small increase in cost per bit.
The whole study was completed in under a week.
What is your application’s design point?
NplusT’s characterization services identify the optimal NAND operating conditions and trade-offs to meet your application targets.
Contact us to scope a study.