What is our primary use case?
Our focus primarily revolves around managing the Hypervisor in the Hyper-Converged Infrastructure. While Sangfor grants us comprehensive control over the Hypervisor, it restricts direct access to VMs, preventing us from making edits or adding features. Instead, it takes responsibility for these actions, requiring us to coordinate sessions or meetings to address any issues. It offers a migration tool that simplifies the process significantly. When clients running on on-premises infrastructure decide to migrate, they can easily download the migration tool onto their VMs, establish network connectivity, and migrate their VMs to our cloud.
What is most valuable?
What stands out to me is the dual-end user interface they provide. The process of creating firewall rules or deploying virtual machines—whether Linux or Windows—is streamlined. Uploading images and configuring VM settings like CPU and RAM are straightforward.
What needs improvement?
While it excels in migrating VMs from on-premises and local infrastructure, it does not provide support for public cloud migrations. Attempting to migrate from public clouds like AWS encountered challenges, and Sangfor's support team clarified that they don't offer assistance for migrations from public cloud environments. It would be much more convenient if the migration tool could be installed directly on the customer's VMs, enabling a smoother migration process to the new infrastructure, with potential restrictions addressed accordingly.
For how long have I used the solution?
I have been using since last year.
What do I think about the stability of the solution?
The reliability has been exceptional, with no breakdowns or disruptions. It offers a seamless experience, and I can confidently select and manage my infrastructure without any concerns about downtime.
What do I think about the scalability of the solution?
It demonstrates scalability, and over the past year, there have been version upgrades from 6.7.0 to the current version, 6.7.9. The changes introduced in the updates offer increased flexibility, which is beneficial. At least five hundred individuals use the service.
How are customer service and support?
The user has faced recurring issues with a customer's VM in a PBX call center solution. One NIC card intermittently disables itself, impacting the system. Despite multiple troubleshooting attempts, including OS changes and NIC card replacements, the problem persisted. While Sangfor's critical support eventually resolved the issue, the prolonged resolution time impacted the overall support experience. Overall, they offer a commendable support structure, involving levels one, two, and three support, with prompt escalation for urgent matters and 24/7 availability. I would rate it eight out of ten.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
In contrast to major cloud providers such as GCP, AWS, Azure, and Alibaba Cloud, Sangfor's portal for service providers is notably user-friendly. With our previous hypervisor, OpenNebula, we faced challenges where the entire infrastructure would go down, resulting in customer complaints.
How was the initial setup?
The initial setup is remarkably straightforward. Installing the operating system on each server and configuring them to form an aggregate cluster can be accomplished effortlessly.
What about the implementation team?
In our role as a cloud service provider with two data centers, we've implemented Sangfor on four to five nodes in one data center, establishing a mirrored site in the second data center for disaster recovery scenarios. As a cloud engineer specializing in digital services, deploying and managing clusters with Sangfor on the cloud is notably straightforward compared to migrating from on-premises infrastructure to major public cloud providers like AWS and GCP. The installation involved configuring connectivity between servers, aggregating resources like CPU, RAM, and storage. We established a private network, granting us easy access to the servers. While I'm not a network specialist, I understand the basic workings of networking. Private subnets with assigned IP pools facilitated seamless server access within the data center. For public access to customers, we incorporated IP pools in the HCI Hypervisor, allowing us to attach a public network to VMs. This ensured customers faced no issues accessing their VMs, whether through SSH, FTP, SFTP, or other protocols.
What's my experience with pricing, setup cost, and licensing?
Its "pay as you grow" model offers cost-effectiveness compared to major cloud providers. However, it may not be the best choice if its costs exceed public clouds. The licensing for our company is structured for a minimum of five years per license. Each license is associated with a node, and if I decide to add another node, I am required to purchase an additional node license.
What other advice do I have?
For entrepreneurs in Europe considering starting a company, it is recommended as a service provider due to its robust competition with VMware and SandForce. While suitable for expanding infrastructure with features like APIs and load balancers, it has drawbacks such as the absence of APIs, Google Drive services, and AI/ML offerings, limiting its appeal for businesses focused on these technologies. It is better suited for middle-sized businesses, with challenges for higher-level businesses scaling in AI. It's essential to evaluate specific business needs before choosing a cloud service provider. Overall, I would rate it seven out of ten.
Which deployment model are you using for this solution?
On-premises