refresh(PIX) - Fotolia
In an age of quick business cycles, where flexibility and agility can provide a competitive edge, businesses in Southeast Asia are turning to converged or hyper-converged infrastructures.
According to the IDC Asia-Pacific C-Suite Barometer 2015 study, converged infrastructure is one of the top three emerging technologies being explored by respondents in the Association of Southeast Asian Nations (Asean) in 2015-2016.
Globally, the hyper-converged integrated systems (HCIS) market is on the rise, and expected to grow from $371.5m in 2014 to nearly $5bn by 2019, according to Gartner.
But currently, hyper-converged infrastructure adoption is still in its infancy, especially in Southeast Asia.
“Within the Asean countries, there is still a need to understand hyper-convergence,” said PK Lim, Asean managing director at storage supplier Nutanix. “[But] we see a growing level of adoption among customers in Indonesia, Malaysia, Singapore and Thailand as they look for ways to simplify IT and reduce costs.”
The idea of a converged infrastructure centres on a hardware-based approach to virtualisation. This means the compute, networking and storage components are all hardware-based and are building blocks in a single chassis. This decoupling of the storage, computing and network software from the underlying infrastructure has resulted in a new set of software-defined infrastructure that runs on industry-standard x86 components.
In contrast to converged platforms that combine components from multiple suppliers, the newer hyper-converged platforms use a software-defined infrastructure on top of physical components. It is typically an all-in-one appliance with all the compute, networking and software from a single supplier.
“If required, the technologies in a converged infrastructure can be separated and used independently. The technologies in a hyper-converged infrastructure, however, are so integrated that they cannot be broken down into separate components,” said Lim.
The demand for these infrastructures is driven by a move from IT infrastructures that are traditionally siloed, to virtualised environments that support cloud computing
“Technology plays a crucial role in enabling the success of the business,” said Cynthia Ho, senior market analyst for enterprise servers at IDC Asia-Pacific. “When business cycles are rapidly shrinking, enterprises do not have the luxury of spending four to six weeks patching together disparate compute, storage and network building blocks, testing and fine-tuning them before the bare metal is up and running. This is where converged solutions come in, with the benefits of agility, simplicity and scalability.”
Today, the hottest segment in converged infrastructures is undoubtedly the hyper-converged appliance, said Matt Oostveen, CTO for Asia-Pacific and Japan at EMC Converged Platforms. “In Asia, the excitement is greater than elsewhere as hyper-converged infrastructure had previously been out of reach for many organisations.”
Leading the way
In Asean, the organisations with mature virtualised deployments are leading adoption.
“We have seen that adoption especially in telecoms, financial services and service providers, as the way they reach their users has changed rapidly. Efficiency and speed of go-to-market are critical for their businesses,” said IDC’s Ho.
Read more about converged and hyper-converged infrastructures
- Market watcher Gartner predicts the hyper-converged market will be worth $5bn by 2019 as integrated systems become a staple of mainstream datacentre environments.
- As datacentre operators focus on converged infrastructure to cut costs, IT professionals should keep their CVs aligned to the workplace.
- The impact of hyper-convergence is far more than the mere aggregation of processing, storage and data protection resources into standardised modular nodes.
Oostveen added that verticals such as mining, oil and gas, retail and healthcare are also strongly adopting these virtualised deployments, due to the distributed nature of their IT architecture and their requirement for edge processing with centralised management.
A hyper-converged setup is typically deployed in an existing enterprise environment to serve a specific project or technical requirement to meet the scale and manageability demanded by enterprise IT administrators, said Lim.
“We see significant interest and awareness among customers that have tremendous pain points, are struggling with a three-tier legacy and want to reduce their datacentre size and power consumption. However, datacentres are slow to adopt converged IT infrastructure due to its all-or-nothing nature.”
An evolving market
While a converged or hyper-converged infrastructure has its benefits, organisations need to be mindful of the considerations and challenges.
“Will the application used be efficiently deployed in a virtual environment, and does the organisation need an alternative disaster recovery plan or data migration plan, given that it is now running on a virtual environment? Organisations need to know which goals to achieve with the converged solution,” said Ho.
“Also, the vendors are still in constant evolution. A good example is IBM’s PureSystems solution. When the x86 component was sold to Lenovo, it affected the existing users’ confidence in the product. The more recent acquisition of EMC by Dell signals more changes ahead.”
The suitability of a converged infrastructure for an organisation would depend on the application/s used, added Ho.
“Not all applications are suitable to be virtualised or run in the cloud. It really depends on the workloads or applications,” cautioned Ho. “The recommended hyper-converged workloads include SAP Hana, virtual desktop infrastructure (VDI) and analytics.”
EMC customers that choose a converged infrastructure tend to use a building block-based system to run their most mission-critical and business-critical applications, with mixed workloads being delivered from the single system.
On the other hand, customers that choose to use hyper-converged infrastructure tend to be discrete single workload use cases such as VDI.
“The biggest difference, though, is not what these systems run, but where they run. The VxRail – EMC’s new hyper-converged appliance – tends to be deployed in remote and regional locations. A great example of this is their use in offshore oil platforms, where systems must be able to operate without the support of an IT team, but be easily managed from a central location,” said EMC Converged Platforms’ Oostveen.
PK Lim, Nutanix
Hyper-converged appliances have economic as well as technical limitations, which mean they are not suitable for all use cases, he added. For example, when the scale point reaches certain thresholds, it is often more cost-effective to deploy on hyper-scaling rack or block systems.
“From a technical standpoint, there are physical limitations to all hyper-converged appliances on the market, where performance degradation begins to impede linear scaling. While hyper-converged infrastructure is highly capable, it is still not best suited to core datacentre workloads – in this scenario the block system would be the preferred converged solution,” said Oostveen.
Before an organisation decides to deploy hyper-convergence technology, it should consider scalability, mobility and security, as well as the best way to adapt it.
“The true power of hyper-converged platforms is that they allow you to get digital products to market faster, while streamlining your expansion, provisioning and management tasks,” said Nutanix’s Lim.
Ultimately, building a converged infrastructure for an organisation would mean replacing and deploying some new servers. It involves being mindful of the considerations and challenges, as well as taking a different perspective to managing the corporate network and storage infrastructure.