Established in 1990 in Indonesia, PT Yakult Indonesia Persada was licensed by Yakult Honsha Co., Ltd., Japan to produce a unique probiotic dairy product made by fermenting a mixture of skimmed milk with a special strain.
- Many of the physical servers were old, at risk of hardware failure, and needed to be retired permanently.
- In the event of hardware error (server failure, disk failure, etc..) a speedy recovery option was mandatory, requiring a high-availability solution.
- The client required a solution that reduced their dependency on power and cooling energy.
- Separation of computing and storage was required with external storage capable of extending the storage capacity.
- A centralized management platform for different hardware components was required.
- Automatic backup features for all the data were mandatory.
Yakult originally had 20 physical servers and 15 VMs running on VMware. Sangfor proposed a customized solution of Sangfor HCI aServer with aSV and aSAN license. We replaced all the old physical servers with only 4 Sangfor HCI servers and migrated the VM with high utilization through the Sangfor platform. In upcoming upgrades Sangfor plans to migrate all VMs from VMware to Sangfor HCI.
Sangfor used four x86 servers and aServer 2200 to build one cluster allowing for potential extension of up to a maximum of 64 servers in one cluster.
Sangfor used two switches to build the storage area network as a redundancy with a 10GB connection. A 10GB port was used as a business port for VM to access the network. A 1GB port was installed as a dedicated management port.
Proposed Migration Plan:
Step 1: Migrate the existing physical servers onto Sangfor HCI. Most of the existing physical servers were old and at high risk of hardware failure which would lead to data loss, significantly impacting business.
Step 2: Migrate all virtual machines running VMware by first managing them using Sangfor HCI on a centralized platform. Then begin a gradual migration of the VMs to Sangfor HCI based on individual VM status beginning with the most utilized VM and working down.