The move to an all-flash data center (DC) will shape the future of infrastructure, helping enterprises harness the power of emerging technologies and applications while also maximising the value of data innovation.
Industries that require high performance and stability — such as finance, telecommunications, and manufacturing — are already turning towards the all-flash DC to leverage its advantages and unlock the potential of their data.
Before we explore some of the real-world success stories of applying the all-flash DC, be sure to check out my earlier posts for an understanding of what the all-flash DC is and how it works.
Financial Industry: Ensuring data security and supporting service innovation
As digitalisation and informatisation sweep across the financial industry, focus has turned to ensuring the high availability of infrastructure, which has high requirements for service response. Financial institutions and organisations whose businesses depend on data have strict demands on the security, continuity, disaster recovery (DR), and backup of critical and sensitive data.
The need to achieve high concurrency and ultra-high peak rates is becoming essential, especially as mobile services increase and their scale expands. More and more organisations require stable, efficient, reliable, and elastic IT infrastructure to achieve business goals and plan strategically.
The all-flash DC provides robust support for scientific and technological innovation by upgrading media, integrating resources, and constructing an architecture that meets the digital development needs of enterprises. It also offers high performance and stability to meet data processing requirements, delivering high input/output operations per second (IOPS), low latency, easy scalability, and high security for key applications and databases, as well as supporting data DR, backup, and archiving.
Success Story: ITAU — Enabling Collaboration and Business Development
ITAU is the largest bank in Latin America and one of the largest in the world, with about 96,000 employees in more than 20 countries. It decided to close its physical branches due to the COVID-19 pandemic, resulting in a sharp increase in the volume of online and mobile banking services. The increased workload caused subpar system efficiency and service system performance across ITAU’s two DCs that were built in 2014, affecting the statistical analysis system (SAS) in particular. This limited the performance of the core debit card service system and made overcoming the storage bottleneck the bank’s top priority.
To address these issues, ITAU adopted the all-flash DC, featuring NVMe, innovative hardware, and intelligent algorithms. Along with providing stable storage performance, the all-flash DC fully unleashes the potential of all-flash. And through RAID-TP, it tolerates the failure of three disks simultaneously with zero service disruption, ensuring the continuity and stability of mission-critical services.
Telecoms: Improving internal management and enabling service innovation
In addition to undergoing digital transformation itself, the telecommunication industry also helps other industries achieve digital development. As technologies such as 5G and cloud computing take hold, telecom operators are facing the pressure of higher internal workloads. At the same time, operators provide value-added services externally, such as smart transportation, smart retail, IoV, gaming and entertainment, and AR/VR applications, requiring fast data access.
Operators’ DCs are therefore expected to provide high-concurrency, low-latency data transmission, and service continuity, which in turn imposes high requirements on the performance and stability of the infrastructure. To cope with service growth and ensure service efficiency, operators are looking at powerful all-flash DCs to accelerate the consolidation and replacement of old devices, modernise storage systems, and maximise their storage investments.
All-flash DCs decouple storage-compute deployment to resolve the reliability weaknesses of coupled storage-compute deployments. They also optimise resources to minimise redundancy and implement elastic expansion of compute and storage resources. And as flash technologies continue to advance — think persistent memory technologies — they not only reduce data access latency but also deliver outstanding performance and enterprise-class storage management functions.
Success Story: A major telco — Ensuring sustainable core services and boosting satisfaction
A major telco’s IT systems faced performance and stability challenges due to higher workloads. To ensure customer service efficiency and quality, the company developed TeleDB, a database based on MySQL. It also systematically reconstructed its IT system architecture to improve database reliability and automate O&M. At that time, the company had three main requirements:
1. High reliability to ensure smooth running of core services covering a wide range of end-users
2. Agility to facilitate the development of new services
3. Cost-effectiveness to save costs by using optimised architectures
Adopting a decoupled storage-compute architecture for its new all-flash DC was an obvious choice, as it ensures the reliability and stable operation of the company’s core services. And by using a large number of NVMe SSDs, it delivers vast improvements in service performance.
By deploying the all-flash DC, the company has not only shortened the failure recovery time from hours to minutes, but also cut energy consumption and freed up floor space in the equipment room. On top of that, the all-flash DC also facilitates the company’s development of new businesses in the era of “Internet+” and mobility thanks to its future-ready architecture.
Manufacturing: Enabling intelligent manufacturing
The key to successful digital transformation of the manufacturing industry lies in properly managing the entire data lifecycle, from collection, storage, and transmission, to presentation, analysis, and optimisation.
In the past, manufacturers have built large-scale application systems but failed to eliminate the information silos that exist between them. As more sensors are used in automated production, and more data needs to be collected, stored, computed, managed, and utilised, new data infrastructure is required to support these growing demands.
To achieve intelligent manufacturing, the manufacturing industry is undergoing digital reform. As part of this, all-flash products are being applied to not only core production systems, but also edge service scenarios as well. All-flash DCs can provide reliable and agile real-time data services for high-end manufacturing, essential in realising intelligent manufacturing.
Take chip product manufacturers as an example. They have offloaded the entire process of circuit design, performance analysis, and IC layout to computers. Logical and physical verification is extremely important, and manufacturers need to ensure that products meet all performance requirements before starting production. The all-flash DC is critical here, as it achieves ultra-fast computation and large-scale storage for things like computer aided design, manufacturing, testing, and computer aided engineering.
Throughout the manufacturing industry, simulation is a crucial step in the manufacturing process. It must be completed as quickly as possible without faults, posing challenges to the performance of infrastructure, especially storage systems.
Success Story: HiSilicon — Ensuring Business Stability and Boosting Satisfaction
HiSilicon is a leading provider of chips and solutions for things like wireless networks, fixed networks, and digital media. It boasts leading IC design and verification technologies, and has an advanced electronic design automation (EDA) platform with sound development processes and specifications. HiSilicon’s EDA data is divided into four categories: public data, project result-related data, project process-related data, and archived data. Simulation accounts for a large proportion of HiSilicon’s workload and therefore has higher requirements for performance.
The four types of EDA data each has specific requirements:
Public data requires high IOPS and high reliability, as it involves more reading and less writing.Project result-related data requires high reliability, as it involves frequent reading and writing.Project process-related data requires high OPS and bandwidth, as it involves frequent reading and writing.Archived data requires remote backup capabilities.
To meet these requirements, HiSilicon adopted the all-flash DC. It uses all-flash NAS storage that helps significantly reduce the time required for simulation. It has also deployed a full SSD-virtualised resource pool, a reliable system, a fully-connected reliable architecture, and a gateway-free active-active solution to ensure its simulation workload operates reliably 24/7. In addition, it employs intelligent QoS to manage multi-user OPS and bandwidth while ensuring the priority of key services.
The future Is All-flash DCs
These are just a few examples of how some industries are adopting the all-flash DC. But as digital transformation touches every industry — be it aviation or healthcare, transportation or mining — the all-flash DC will become the go-to solution for all who want fast access to the large-scale storage of their data.
The all-flash DC will shape the future of infrastructure. Indeed, its effects will reverberate across all industries as they undergo digital reform, bringing near-immediate responsiveness to mission-critical data and workloads.
In my next and final post of this series, I’ll wrap up the benefits of the all-flash DC and look at why it might be the right choice for you.
This post is adapted from the Huawei-sponsored IDC White Paper Moving Towards an All-Flash Data Center Era to Accelerate Digital Transformation.