Make Your Digital Transformation

Simpler and Secure

ENTERPRISE
STORAGE
MADE EASY

www.sangfor.com

marketing@sangfor.com
 What is Enterprise Storage?

Enterprise storage is a centralized repository for business information that provides common
data management, protection, and sharing functions through connections to server systems
and/or clients.

Compared to desktop or consumer-grade storage, enterprise storage systems should be

scalable for workloads of hundreds of terabytes (TB) or even petabytes (PB) without relying on
excessive cabling or the creation of subsystems. An enterprise storage system should be
scalable without limitations on connectivity and be able to support multiple platforms/protocols.

Approaches to Traditional Enterprise Storage

SAN DAS NAS

A storage area network (SAN) is a Direct-attached storage (DAS) is a Network-attached storage (NAS)
dedicated, high-performance storage disk drive chassis connected enables multiple servers or client
network designed for servers and directly, either inside or outside (in devices and users to access data
is separate from the user data a storage enclosure), to a SAN/ from a central pool of disk storage.
network. It interconnects pools of NAS controller to expand the Users access the shared storage
disks or solid-state storage and capacity of existing storage of NAS over an IP/Ethernet
shares block storage with multiple systems. connection. It is characterized by
servers. its ease of access, low cost, and
large capacity.

Pros

Consistent Performance
Purpose-built and optimized software and hardware that work smoothly

Reduced Footprint Reliability

With optimized software and hardware, typically in a 2U Due to the closed hardware and software, support
chassis with an initial storage pool and dual controllers, and maintenance are provided by the vendor or
you can scale up with additional DAS disk enclosures certified personnel to ensure quality

ENTERPRISE STORAGE MADE EASY 01

 Cons

High initial costs

Limited Scalability: Vendor Lock-in:

The hardware model limits Supports only specified proprietary
scalability, and upgrade options are software and hardware.
usually limited to Lift and Shift only.

Complex Management: High Total Cost of Ownership (TCO):

Purpose-built software and some Requires specific consumables,
operations involve command lines that accessories, upgrade parts, and skill
require storage experts to handle. sets only available from the vendor.

Major Enterprise Storage Trends

Newer enterprise storage approaches and technologies that have trended upward
over the last several years include cloud, hyper-converged, and flash storage, such as
non-volatile memory express (NVMe).

Storage for containers is also gaining traction, and enterprise storage based on
disaggregated and composable infrastructure concepts is one of the trending
technologies.

Disaggregated and Composable Storage

Some observers view disaggregated and composable infrastructures as the next stages in the evolution of hyper-converged
infrastructure (HCI). They retain the benefits of hyper-convergence, such as physically combining computing resources into an
easily scalable framework via nodes, while making it easier to add storage and compute hardware independently.

Disaggregation is accomplished by separating these computing components into discrete pools

of CPU, cache, fabric, memory, and storage resources that can be served on demand to specific
applications. It combines these individual resources at the hardware level and assembles them at
the software level using application programming interfaces (APIs).

Composable infrastructure creates, or composes, consumable sets of hardware resources and

combines them into a virtualized whole. So, instead of disaggregating these components into
virtual resources at the hardware level and then serving them up individually using software,
composable infrastructure combines these singular pools of resources into a unified virtual
infrastructure unit, operated by a software entity.

ENTERPRISE STORAGE MADE EASY 02

 Hyper-Converged Storage

Hyper-converged storage integrates storage, compute, and sometimes networking

into a single unit. It adds a virtualization layer into the mix, with storage managed
through the hypervisor. That makes hyper-convergence a type of software-defined
storage that enables all the storage resources in each node to be pooled across a
cluster, giving administrators more control of storage provisioning in a virtual server
environment.

This gives hyper-converged storage a much higher degree of horizontal scalability

because you simply plug a new node into your infrastructure to expand storage and
compute resources.

A hyper-converged infrastructure (HCI) is different from a converged infrastructure (CI), which bundles the traditional SAN
components of storage, compute, and networking into a preconfigured stock-keeping unit (SKU) that is sized and tuned before sale.
HCI converges these capabilities into a single appliance that can be clustered with other similar nodes. CI is a staple of enterprise
storage, while HCI began mainly as an efficient way to run heavily virtualized loads, such as virtual desktop infrastructure (VDI). HCI
has since made its way into the enterprise, and now, HCI clusters typically run multiple workloads.

Sangfor HCI’s aSAN storage component is equipped with local copy technology as a storage optimization feature. A full copy of the
virtual disk image is resident in the same physical HCI node that the virtual machine (VM) is running on. This helps to eliminate the
network overhead of disk I/O latency, making it suitable for demanding applications or other mission-critical systems with artificial
intelligence (AI), online transaction processing (OLTP), and online analytical processing (OLAP).

Container Storage

Unlike standard monolithic applications, containerized applications can consist of hundreds or thousands of individual related
containers, each hosting isolated and separate scalable processing modules of the overall application process. Containers were
designed to allow users to easily develop and deploy stateless microservice layers of an application as a type of agile middleware,
with no persistent data storage required.

One reason containerized application design has grown in popularity with enterprises over the last several years is the design’s
support for agile application development and deployment. Containers can quickly scale up as needed in a production environment
and then be decommissioned when no longer needed. It is the efficiency, scalability, agility, cloud-friendliness, and lower costs of
this approach that have led enterprises to turn to container architectures for purposes beyond microservices.

As a result, container vendors, such as Docker and Kubernetes, have started offering
some level of persistent data storage support to containers. In addition, startups such
as Portworx have addressed this issue—often more thoroughly—by enabling container
storage volumes to move with their containers.

Container data, like all enterprise data, needs protection. However, containers, by their
nature, cannot directly interface with standard backup applications due to a number of
factors. They require an underlying orchestration platform at the storage or host layer.
A number of storage vendors—including Asigra, Blockbridge Networks, Commvault,
NetApp, Nimble Storage, and Pure Storage—have stepped up and enabled enterprises
to back up container data by offering this orchestration layer.

ENTERPRISE STORAGE MADE EASY 03

 Cloud Storage

Enterprise cloud storage consists of storage capacity offered by a public cloud service provider (CSP)
on a subscription basis. There are hyperscale public cloud storage providers and a host of regional
providers. Enterprises often turn to cloud storage to reduce or eliminate excessive on-premises
storage costs, reduce storage management complexity, and upgrade data center infrastructure.

There are also systems integrators and managed cloud service providers offering localized cloud
services, and many traditional software companies (ISVs & ASPs) now offer cloud storage services for
their applications and services. Sangfor advocates for users and partners to incorporate enterprise
cloud storage into their hybrid architectures, as internet service providers (ISPs) and CSP vendors have
integrated Storage-as-a-service (STaaS) into their Everything-as-a-Service (XaaS) service portfolios.

Introducing Sangfor aStor – Enterprise Distributed Storage

Sangfor aStor is available in two forms:

Converged Storage Standalone Software-Defined, Scale-Out, Unified Storage

Provides converged storage modules for Since 2018, aStor has delivered projects over the 100
Sangfor VDI and Sangfor HCI. This form of PB level, and the cumulative amount of stored data has
storage has delivered more than 20,000 exceeded 4,000 PB, supporting core business
customer cases to date. operations with zero data loss.

Gov Cloud/Local Cloud Healthcare Enterprise Media

Solution

Cloud Massive High Bleclronic Backup Unified Anti-Virus Large

IOT
Storage Unstructured Performance PASC Health DR Storage Anti- Volume Archives
R&D
Services Data DB Records Archives Pool Ransomware Media
Application

VM/DB Docker windows File APP Hadoop

DB/VM/ Windows Linux

Container Applications Big Data apps
CMP File Sharing File Sharing

iSCSI FC Cinder CSI CIFS FTP NFS S3 SWIFT HDFS

SAN NAS Object HDFS

Software

Sangfor aStor

All Flash Resource Pool Hybrid Disks Resource Pool

Hardware

Generic server cluster (scale-out)

ENTERPRISE STORAGE MADE EASY 04

 Product Values

Unified Management
Traditional SAN storage often requires high-performance array storage for databases and
virtualization environments, and NAS storage for file-based data, such as IoT data, images, and
videos. This requires the maintenance of multiple sets of storage devices at the same time. In
comparison, Sangfor aStor only requires three nodes to form a unified storage resource pool,
providing SAN (block), NAS (file), and object storage services. One unified storage system
hosts all types of business data with enterprise-grade performance and high availability.

Evergreen Capacity and Performance Growth

In terms of scalability, traditional storage is analogous to a steam engine; it slows down when
additional carriages are added to expand capacity. In contrast, aStor is likened to a high-speed
train where every carriage is equipped with its own engine, ensuring consistent performance
with capacity expansion. As aStor is distributed (scale-out) storage, it achieves linear growth in
capacity and performance compared to traditional storage and tolerates node-level failures
and multi-disk failures in terms of reliability.

Benefits of Sangfor aStor

Cost-effective Secure and Reliable

Through research and development and innovation in software aStor adopts erasure coding redundancy technology to ensure
technology, aStor provides large storage capacity, reaching up data reliability and improve the utilization rate of storage space.
to 10 billion small files. It delivers a data compression ratio of Additionally, the system is built in with anti-virus (TBA) and
more than 50%, and a single node supports more than 1GB/s in WORM (Write Once, Read Many; immutable) features to counter
throughput and more than 100,000 IOPS. threats like viruses and ransomware.

Openness Scalable

Support third-party software integration, including backup, The distributed architecture forms a storage cluster of
cloud storage, and other functions. It also supports general-purpose x86 servers, which supports online
scenario-based applications and allows data to be shared horizontal expansion of nodes, and performance
securely through NFS, CIFS, FTP, HDFS, and other interfaces. increases linearly with capacity.

ENTERPRISE STORAGE MADE EASY 05

 For Management

Cost-effective
aStor delivers higher storage performance with a lower hardware investment. Only the
configuration of mixed flash can compare to the performance of full flash (certified by third-party
test reports). The use of general-purpose x86 servers allows for smooth hardware upgrades,
lower maintenance costs, and lower outdated equipment replacement costs.

Unified Storage Management

aStor only requires three nodes to provide 3-in-1 unified storage (SAN, NAS, and Object) to cater
to the needs of most scenarios.

For IT

aStor provides intelligent O&M capabilities to ensure the high reliability of storage platforms.
These include automatic data reconstruction after a fault, isolation of sub-health hard disks, and
proactive bad sector scanning and repair. It also improves the utilization rate of storage
resources through automatic data balancing mechanisms. No manual intervention is required in
the whole process, alleviating the O&M workload of administrators.

Comparison at a Glance

Software Defined, Traditional Cloud Software

Storage Scale-out, Unified
Storage Storage Defined Storage
Storage

Dell EMC, IBM, HDS AWS, Azure, GCP, Red Hat Gluster,
Sangfor aStor
Market Player NetApp, HPE, Huawei, Alibaba Cloud, Huawei VMware vSAN, NetApp
H3C Cloud ONTAP Select

Physical Appliance Yes Yes (mandatory) No No

Propriety System No, general x86 Yes (cost concern) Yes No, general x86 server

Software Defined with

server No (H/W lock-in) No Yes
General X86 Server

Software License Only Yes No No Yes

Deploy On-premises Yes Yes No, only caching appliance Yes

ENTERPRISE STORAGE MADE EASY 06

 Software Defined, Traditional Cloud Software
Storage Scale-out, Unified
Storage Storage Defined Storage
Storage

Deploy to Local Cloud Yes Possible (ROI concern) No Yes

No, Performance
Scale-out Yes is limited by the Yes Yes
controllers

Optimal SAN: SAN: RAID 10 (equal to

Redundant paths (RP) RP2) / RAID 5,6 (less Mostly EC (Erasure
Storage Virtualization Block Storage: 2
with 2/3 copies performance) Coding), not optimal
Technology copies
Optimal NAS: Erasure NAS: Additional H/W or for SAN
coding (EC) 4/8+2 S/W on top of SAN

Physically Isolated No, Shared LUN from

Yes No Yes
Cluster the controllers

Physical Node Fault

Yes No Yes Yes
Tolerance

Guaranteed
Yes Yes No Yes
Performance

Unified Storage Yes No No Partial

Centralized Storage Yes Partial Yes No

Management

Block Storage Yes Yes Separated service Partial

Partial, mostly
File Storage Yes separated models or Separated service Yes
additional file head

Object Storage Yes Separated models Separated service Yes

Compression Yes Yes Yes Yes

Deduplication &
No Yes Yes Partial
Encryption

Anti-Virus TBA 2024 No No No

WORM (Immutable) Available Partial Partial Partial

Unified Management Yes No No No

with HCI

For further information and Enterprise Distributed Storage design or pricing information, please contact your local
Sangfor office and/or local partners.

ENTERPRISE STORAGE MADE EASY 07

 Make Your Digital Transformation
Simpler and Secure

https://twitter.com/SANGFOR

https://www.linkedin.com/company/sangfor-technologies

https://www.facebook.com/Sangfor

https://www.instagram.com/sangfortechnologies/

https://www.youtube.com/user/SangforTechnologies

Sales: sales@sangfor.com

Marketing: marketing@sangfor.com

Global Service Center: +60 12711 7129 (or 7511)

www.sangfor.com