Huawei E9000 Blade Server

Featuring high energy efficiency, outstanding performance, high scalability and virtualization performance, E series blade server is an ideal choice for customers' growing business by consolidating computing, switching, storage, IOs and management subsystems into one single chassis.

Huawei E9000 Blade Server Chassis

The FusionServer E9000 converged-architecture blade server enables convergence of computing, storage, networking, and management.

It provides a powerful platform for high-end carrier or enterprise applications, and is ideal for data centers, virtualization, mission-critical services, and high-performance computing.

Provides 16 slots in a 12U chassis that includes redundant Power Supply Units (PSUs), heat-dissipation modules, management modules, and switch modules.

The Huawei E9000 server can be installed in a standard 19-inch rack at a depth of at least 1m. It is available with AC or DC power setups to suit data center requirements

Huawei E9000 Front View Huawei E9000 Rear View

 

Hightlight

Specification

Form Factor

12U

Blade

16 half-width slots or 8 full-width slots

Supports flexible configurations of full-width and half-width nodes

Switch Modules

4 slots for Huawei CX series switch modules provide a midplane switching capability of 15.6 Tbit/s

CX110 GE switch module: 12 x GE + 4 x 10 GE uplink, 32 x GE downlink

CX111 GE switch module: 12 x GE + 4 x 10 GE uplink, 32 x GE downlink

CX116 GE pass-through module: 32 x GE uplink, 32 x GE downlink

CX210 8G FC switch module: 8 x 8G FC uplink, 16 x 8G FC downlink

CX220 16G FC Switch Module: 8 x 16G FC uplink, 16 x 16G FC downlink

CX310 10 GE switch module: 16 x 10 GE uplink, 32 x 10 GE downlink

CX311 10 GE/FCoE converged switch module: 16 x 10 GE + 8 x 8G FC uplink, 32 x 10 GE downlink

CX317 10 GE pass through module: 32 x 10 GE uplink, 32 x 10 GE downlink

CX318 10 GE pass through module: 32 x 10 GE uplink, 32 x 10 GE downlink

CX611 InfiniBand switch module (QDR 40 Gbit/s, FDR 56 Gbit/s): 18 x QDR/FDR uplink, 16 x QDR/FDR downlink

CX710 40 GE switch module: 8 x 40 GE uplink, 16 x 40 GE downlink

CX912 multi-plane switch module; supports FC ports: 16 x 10 GE + 8 x 8G FC uplink, 32 x 10 GE + 16 x 8G FC downlink

CX915 multi-plane switch module: 4 x 10 GE + 12 x GE + 8 x 8G FC uplink, 32 x GE + 16 x 8G FC downlink

PSU

AC/DC PSU: maximum six 3,000W/2,000W AC or six 2,500W DC hot-swappable PSUs in N+N or N+M redundancy mode

Fan Modules

Provides 14 hot-swappable fan modules in N+1 redundancy mode

Management

  • Uses HMMs in 1+1 redundancy mode, supports standard management interfaces, such as SNMP and IPMI, manages computing and switching devices in a unified manner, and provides comprehensive management functions, such as a GUI, remote KVM, virtual media, SOL, remote control, hardware monitoring, intelligent power supply, and historical power consumption
  • Provides value-added management features, such as multi-frame cascading management, graphic network configuration, and stateless computing

Power Supply

110V/220V AC or –48V DC

Operating

Temperature

5°C to 40°C

Dimensions

Height: 530 mm (20.87 in.)

Width: 442 mm (17.40 in.)

Depth: 840 mm (33.07 in.)

 

 

E9000 Compute Nodes

 

Model Picture Width brief introduction
CH121 Ch121 appearance half width 2-socket E5-2600 V3/V4 family contain up to two 22-core processors to power each compute node, 24 DDR4 DIMMs, providing a maximum memory capacity of 1.5 TB
CH140 CH140 appearance Two 2-socket, twin compute nodes in a half-width slot Supports the full series of Intel® Xeon® E5-2600 V3/V4 processors to deliver up to 2 x 22-core of computing capacity; a half-width slot supports two small slots in two layers for installing two independent 2-socket compute node
CH220 CH220 IO expansion module appearance Full-width 2-socket I/O Expansion Compute Node  E5-2600 v3/v4 series delivers up to two 22-core processors on each node, Six standard PCIe card expansion slots for various PCIe configuration modes provide the best possible PCIe expansion capability on a single compute node
CH221 CH221 IO expansion compute node appearance Full-width 2-socket I/O Expansion Compute Node Supports the full series of Intel® Xeon® E5-2600 series quad-core, six-core, and eight-core processors to deliver up to 2*eight-core 2.9 GHz of computing power. Provides 24 DIMMs at 1.5 times the usual height. These apply mainstream granules and double the memory capacity to up to 768 GB DDR3 (providing the highest cost-efficiency for large-memory applications). Provides 4 PCIe x8 full-height half-length slots for installing 4 GPUs or PCIe SSDs, providing the best possible PCIe expansion capability on a single compute node
CH222 CH222 Storage expansion compute node appearance Full-width 2-socket Storage Expansion  compute node The CH222 provides superior computing performance and a large storage capacity. Featuring Intel® Xeon® E5-2600 series processors (up to 135 W) and support for up to 24 DIMM slots, 15 x 2.5" SAS/SATA HDD or SDD hard disks, and a 1 GB RAID cache, the CH222 is suitable for big-data analysis and processing applications that require large storage capacity and high computing performance, such as videos, searches, and biological sciences.
CH225 CH226 storage expansion compute node Full-width 2-socket All-flash Storage Expansion  compute node Supports up to 12 x 2.5-inch NVMe SSDs and 2 x 2.5-inch SAS/SATA HDDs or SSDs, delivering up to 9,600,000 IOPS. Delivers industry-leading NVMe storage capacity per node, Supports the full series and all specifications of Intel® Xeon® E5-2600 V3/V4 processors. Provides 2 x 22-core processors to deliver superior computing capability
CH226 CH226 storage expansion compute node Full-width 2-socket Storage Expansion  compute node Intel® Xeon® E5-2600 V3/V4 processors to deliver up to 2 x 22-core of computing capacity, 24 DDR4 DIMMs, Supports 6 x 3.5 in. SAS/SATA HDDs and 2 x 2.5 in. SAS/SATA HDDs or SSDs, which provide the highest storage capacity on a single node
CH240 CH226 storage expansion compute node Full-width 4-socket compute node Supports Intel® Xeon® E5-4600 series processors, with up to 8 cores each, a 20 MB L3 cache, 8 GT/s QPI, hyper-threading, and Turbo acceleration technology. Provides 48 DIMMs at 1.5 times the usual height. These apply mainstream granules and provide the memory capacity of up to 1.5 TB DDR3 (providing the highest cost-efficiency for large-memory applications). Supports up to 8 x 2.5" SAS/SATA HDDs or SSDs, or SSDs (twice as many hard disks as competitors' 4-socket blade servers).
CH242 CH226 storage expansion compute node Full-width 4-socket compute node Supports new-generation Intel® Xeon® E7-4800 v2 series 8-core, 10-core, 12-core, and 15-core high-performance processors based on the Brickland platform, and provides outstanding computing performance with up to four 15-core 2.8 GHz processors. 4-socket compute node installed in a full-width slot, supports up to a maximum of 32 DDR3 DIMMs at 1.5 times the usual height that apply to mainstream granules, addressing large-memory-capacity applications at low cost. Supports a maximum of eight 2.5-inch SAS HDDs, SATA HDDs, or SSDs, and provides a maximum of 3 PCIe slots for standard PCIe cards.

 

Comparison of Huawei E9000 and competitors' products

Key Competitiveness

Key feature

HP C7000 G8

IBM FlexSystem

Cisco UCS G3

Huawei E9000

Engineering capability

heat dissipation for front ,rear slots and power supply

1200W/100W

1200W/100W

1200W/100W

1400W/280W

Computing capability

Computing density

32

56

16

64

Memory capacity

Memory capacity per chassis (mainstream granules)

8.2T

10.8T

6.14T

12.3T

Storage capability

2.5" disk quantity per slot

12

12

12

15

Computing and storage converged

External SAN

Internal SAN

External SAN

External SAN

Internal DS

network

Backplane Bandwidth

6.144Tbps

7.168Tbps

1.28Tbps

15.6Tbps

10GE/FCoE/8G FC/IB switch

Y

Y

Y

Y

40GE

Y

N

Y

Y

RoCE

N

Y

N

Y

16G FC

N

Y

N

N

PCIe expansion

PCIe resources expansion

N

2*PCIe 8x+2*PCIe 16x

N

4*PCIe 8x

or 2*PCIe 16x

Running environment

Ambient temperature

5°C~35°C

5°C~40°C

5°C~35°C

5°C~40°C

Computing Density Comparison

HUAWEI

E9000

DELL

M1000e

IBM

FlexSystem

SuperMicro

SuperBlade

Quantity of CPUs/Chassis

64

32

32

40

Single node maintenance

Yes

Yes

No

No

Quantity of memory channel

8

6

6

8

Computing Density Comparison

HUAWEI

E9000

DELL

M1000e

IBM

FlexSystem

Quantity of HDDs per slot

15

7

12

Quantity of HDDs per chassis

120

56

84

Switch Performance Comparison

HUAWEI

E9000

DELL

M1000e

IBM

FlexSystem

HP

C7000

Backplane bandwidth

15.6T

8.96T

6.27T

7.168T

Quantity of 10GE

128

48

88

64

 

Comparison Between Huawei E9000 and HP C7000/C3000

 

Key advantage of Huawei E9000

Huawei FusionServer E9000 (E9000 for short) is a new-generation blade server designed for cloud computing and big data applications based on comprehensive analysis of HP's C7000 (launched in 2007), Cisco's UCS5108 (launched in 2009) and VCE Vblock (launched in 2010).

Huawei Advantage Customer Benefits Competitor' defense and Huawei's HTB Strategy
No. 1 computing density in the industry. Number of processors in a 1U chassis:
● Huawei E9000: 5.3/U
● HP C7000: 3.2/U
● HP C3000: 2.7/U
E9000 provides high computing density and performance. Competitors' defense:
HP C7000 servers are transformed from CPU stacking to CPU+GPU heterogeneous computing.

Huawei HTB strategy:
● The number of GPU nodes occupies only 20% in HC cluster. Heterogeneous computing has limited application.
● The E9000 server (configured with CH220 IO nodes) supports CPU+GPU.
No. 1 storage density in the industry. Number of 2.5" hard disks in a 1U chassis:
 ● Huawei E9000: 12/U
● HP C7000: 9.6/U
● HP C3000: 8/U
High storage density ensures high concurrent read and write performance. The concurrent read and write performance is determined by PCIe bandwidth or hard disk IOPS upper limit. Competitors' defense:  HP D2200sb DAS nodes can be combined with adjacent blades, while Vendor H CH222 nodes cannot.

Huawei's HTB strategy: A D2200sb combined with two adjacent blades provides up to 12 hard disks. However, the D2200sb cannot be combined with E7 4-socket blades. A 2-socket blade of E9000 provides 15 hard disks, and a 4-socket blade provides 8 hard disks. Vendor H's C7000 servers do not support PCIe SSD cards, while E9000 storage nodes (except CH140 and CH240) support PCIe SSD cards.
No. 1 switching capability in the industry. Design bandwidth of the backplane:
● Huawei E9000: 15.6 Tbit/s
●HP  C7000 platinum: 6.144 Tbit/s
The E9000 backplane supports 100GE and IB EDR, maximizing return on investment (ROI). Competitors' defense: The E9000 provides four switch slots, while the HP C7000 provides 8 switch slots.

Huawei's HTB strategy: The E9000 provides multi-plane switch modules. Two switch planes are created in a slot. The E9000 supports hybrid networking of Ethernet +FC SAN, IB, and FCoE modules. Except IB, all E9000 switch modules support layer3 switching, while Vendor H's servers support only layer 2 switching.
No. 1 engineering capability in the industry. Power supply and heat dissipation capability of a full-width slot:
● Huawei E9000:1400 W/40°C
●HP C7000:1200 W/35°C
This feature improves system reliability and availability and reduces the PUE of data centers. Competitors' defense: The temperature of 77% data centers is set to 18°C to 21°C. It is seldom for servers to operate at 40°C. In addition, the servers operating at 40°C consume more power. The E9000 servers do not support three-phase power.

Huawei's HTB strategy: ● According to statistics, 7% power consumption is generated to reduce the temperature by 1°C. Nowadays, the cooling devices account for more than 40% of the devices in data centers. Overcooling causes waste of energy. If the data center temperature increases by 5°C in the world, the total carbon dioxide emission will be reduced by 1.7 million tons and electricity of 24.3 billion degrees will be saved. The E9000 servers use intelligent power management fan modules and orthogonal air duct design, which consumes 5% less power at the same CPU load. ● The E9000 servers support three-phase power supply converted by PDUs.

 

Key advantage of HP C7000/C3000

Claimed Advantage Customer Benefits Competitor' defense and Huawei's HTB Strategy
10U chassis
● HP C7000: 4 C7000s in a cabinet
● Huawei E9000: 3 E9000s in a cabinet
The high-density design saves floor space. ● The power distribution for a cabinet in 95% data centers is 6 kW to 10 kW, which supports one high-end blade or two to three low-end blades.
● A 42U cabinet can hold three 12U E9000 servers, leaving 6U space for other devices.
Supports 8 interconnected switch modules.
 ● HP C7000: provides 8 slots for switch modules
 ● Huawei E9000: provides 4 slots for switch modules
The C7000 servers support a wide variety of switch modules, allowing flexible networking. ● According Gartner statistics, data center converged networking will be the mainstream networking mode in the near future.
● The E9000 servers support FCoE networking as well as multi-plane networking, offering flexible configuration and deployment.
Compatible with the blade servers of the same vendor.
● HP C7000: supports x86/AMD/work station/Unix blades/NAS gateway/blade tape drive/SAS switch modules
● Huawei E9000: supports only x86 blades.
High compatibility help maximize ROI of customers. x86 servers account for 90% in the server market. The market share of non-x86 servers is decreasing.
● The E9000 CH220/221 nodes support work stations.
● The demand for blade tape drivers is less than 1%. If required, the E9000 server can meet customer requirements by using an external tape drive. The use of SAS switch modules and JBOD causes the similar cost as the IP SAN but offers poorer reliability than IP SAN. If NAS gateway blades are used, the servers must have WSS installed on compute blades and connect to vendor H's SAN storage. This configuration causes high costs and low reliability. Customers can achieve the same purpose by using an external storage
Provides an LCD on the front panel.
● HP C7000: LCD
● Huawei E9000: touchscreen
The LCD helps maintenance personnel learn about the server status in real time. The Huawei E9000 server provides a touchscreen.
Implements unified management.
● HP C7000: OneView
● Huawei E9000: eSight
OneView implements unified management of computing, storage, and network resources, simplifying management and increasing O&M efficiency. ●Huawei eSight implements unified management of servers.
● Huawei eSight and E9000 internal management modules support stateless computing, implement plug-and-play of E9000 servers in site deployment, equipment migration, and parts replacement.
Provides high power efficiency and optimal heat dissipation.
● Huawei E9000: 1400 W/40°C
● HP C7000: 1200 W/35°C
This feature helps reduce power costs ●If HP C7000 servers use common blades with high power-consuming CPUs, at least 4 to 12 memory modules must be configured.
● The E9000 servers support 1400 W full-width and 700 W half-width nodes, E5-2600 v2 and E7-4800/8800 v2 series processors, and K1/K2/K2000/K4000 /Intel Phi PCIe cards.

 

Comparison Between Huawei E9000 and Cisco UCS5108

Huawei FusionServer E9000 (E9000 for short) is a new-generation blade server designed for cloud computing and big data applications based on comprehensive analysis of HPE's C7000 (launched in 2007), Cisco's UCS5108 (launched in 2009) and VCE Vblock (launched in 2010).

Huawei Advantage Customer Benefits Competitor' defense and Huawei's HTB Strategy
No. 1 computing density in the industry. Number of processors in a 1U chassis:
● E9000: 5.3/U
● UCS: 2.7/U
E9000 provides high computing density and performance. Competitors' defense: The CH140 compute node of E9000 provides only 8 memory slots and can be configured with a maximum of 24 memory modules. It offers poor memory performance.

Huawei's HTB strategy:
● The memory performance is determined by the communication bandwidth between the memory and the CPU. The memory bandwidth is determined by the maximum number of memory channels and memory frequency. The IvyBridge EP platform supports up to 8 memory channels. The CH140 provides 8-channel memory and supports memory modules of the highest frequency. It provides high read and write performance.
● The top two drawbacks of USC servers in clusters are long delay and poor CPU performance.
No. 1 storage density in the industry. Number of 2.5" hard disks in a 1U chassis:
● E9000: 12/U
● UCS: 2.7/U
High storage density allows large local storage space and high concurrent read and write performance. The concurrent read and write performance is determined by PCIe bandwidth or hard disk IOPS upper limit. Competitors' defense: The CH222 storage nodes of E9000 cannot provide shared storage resources to other blades.

Huawei's HTB strategy: The CH222 running Vendor H DSware software can group the local 15 hard disks as a virtual storage resource pool to support more servers. It implements the SAN storage functions. The DSware computing and storage convergence architecture allows a cabinet to provide 1000 Gbit/s (SAN 128 Gbit/s) concurrent bandwidth and 96 Gbit/s SSD (SAN 12 Gbit/s SAS) throughput.
No. 1 switching capability in the industry. Design bandwidth of the backplane:
● E9000: 15.6 Tbit/s
● UCS: 1.28 Tbit/s
The E9000 backplane supports 100GE and IB EDR, maximizing return on investment (ROI). Competitors' defense: The multi-plane networking of E9000 causes complicated deployment and management. The UCS single switching plane implements convergence of services and management as well as convergence of switching and FC SAN, simplifying deployment.

Huawei's HTB strategy: The UCS servers build a unified computing, storage, and management network based on FCoE Fabric Interconnect. However, the networking does not support IB interconnection. The E9000 servers support multi-plane networking as well as FCoE networking. Vendor H eSight implements unified management of servers.
No. 1 engineering capability in the industry. Power supply and heat dissipation capability of a full-width slot:
● E9000: 1400 W/40°C
● UCS: 1200 W/35°C
This feature improves system reliability and availability and reduces the PUE of data centers. Competitors' defense: The temperature of 77% data centers is set to 18°C to 21°C. It is seldom for servers to operate at 40°C. In addition, the servers operating at 40°C consume more power.

Huawei's HTB strategy:
● According to statistics, 7% power consumption is generated to reduce the temperature by 1°C. Nowadays, the cooling devices account for more than 40% of the devices in data centers. Overcooling causes waste of energy. If the data center temperature increases by 5°C in the world, the total carbon dioxide emission will be reduced by 1.7 million tons and electricity of 24.3 billion degrees will be saved. The E9000 servers use intelligent power management fan modules and orthogonal air duct design, which consumes 5% less power at the same CPU load.
● A fire was caught due to poor heat dissipation of UCS servers.

 

Key advantage of Cisco UCS5108

Claimed Advantage Customer Benefits Competitor' defense and Huawei's HTB Strategy
Unified computing Provides stateless computing Configuration drift among blade and rack servers Plug-and-play servers reduce deployment time. ● The E9000 servers support stateless computing and allow plug-and-play of blades in initial deployment, system migration, and parts replacement.

● The UCS5108 servers offer poor resource expansion. The E9000 servers can be configured with CH222 storage nodes and CH220/221 I/O expansion nodes, which offer more advantages in big data and VDI applications.
Unified network Use UCS 6200 FI series FCoE switches to unify LAN /SAN and management planes. Use FEX 2200 series I/O cards to replace switch modules Unified network simplifies networking and costs. One set of FI switch performs switching and management of a maximum of 160 blade and rack servers. The subsequent expansion cost is low. ● The networking mode restricts the UCS5108 servers in convergence and virtualization applications.
● The UCS5108 servers do not support IB interconnection, and the chassis does not support evolution to 100GE.
● The UCS servers use private network protocols.
● The E9000 servers offer lower expansion costs than the UCS5108 servers. E9000 servers, CX310 switches, and external FCoE switches with an uplink convergence ratio (blades to external switches) of 1:4 to a unified LAN+SAN network.
Unified management The FI built-in UCS Manager implements unified management of servers, LAN, and SAN. The cost-efficient FEX cards implement management and I/O connection of subracks and FI switches. One set of FI can manage up to 160 blade and rack servers. Unified management reduces O&M costs. Huawei eSight implements unified management of servers, LAN, and SAN device

 

Huawei E9000 Application Scenarios

 

Enterprise Datacenter Application

Huawei E9000 Enterprise datacenter application 

Typical Configuration

E9000 chassis
CH121 2-socket compute node (for web, middleware, and application VM deployment)
CH220 resource expansion 2-socket compute node
CH222 storage expansion 2-socket compute node
CH240 4-socket compute node (database physical machine deployment)
Large-capacity SSD cards with application acceleration for email, ERP, CRM, databases, and BI
CX110 and CX311 switch modules
GE, 10GE, FCoE, and FC switching, providing a converged network to improve management convenience, flexibility, and scalability.

VDI (Virtual Desktop Infrastructure) in a box

Huawei E9000 VDI application 

Typical Configuration

E9000 chassis
CH222 storage expansion 2-socket compute node
CX110 switch module
PCIe SSD card

Features

Supports 128 processor cores, 6 TB memory capacity, and up to 128 TB raw storage capacity.
Supports 500 (heavy) to 1000 (light) virtual desktop infrastructure (VDI) users.
Supports deployment of an OS and hot applications on PCIe SSD cards, which reduces the VDI startup peak, increases IOPS by 800%, and reduces costs by 20%.
Supports the disk drawer and hot-swappable disks to reduce the time required to rebuild data during disk replacement.
Supports the virtual block storage (VBS) distributed harsh table (DHT) ring to protect shared disk storage in cross-server mode.

 

MPP Architecture OLAP for BI, DW, and In-memory Database

Huawei E9000 MPP architecture 

Typical Configuration

E9000 chassis
CH222 storage expansion 2-socket compute node
CH240 4-socket compute node (1.5 TB memory capacity)
PCIe SSD card
CX311 switch module

Features

Provides 2-socket 24-DIMM computing capability per node. Supports 15 disks per node.
A 4-socket 48-DIMM analysis node provides 1.5 TB memory capacity to support the memory database and real-time analysis.
Each node provides service bandwidth of 20GE. The switch latency is in ns, which removes the IO bottleneck during the ETL process.
The large-capacity PCIe SSD card is used for the index and hot data cache, which greatly improves the IOPS of the parallel database.

 

Unstructured Databse Application Hadoop, Hbase Application

Huawei E9000 unstructured Database Application 

Typical Configuration

E9000 chassis
CH121 2-socket compute node
CH222 2-socket storage expansion compute node (for web, middleware, application VM deployment)
Hadoop hardware compression card
CX110 switch module

Features

A NameNode 2-socket compute node can provide a cost-effective large memory with up to 768 GB memory capacity.
A DataNode 2-socket compute node supports 15 x 2.5" disks, and provides 40GE service bandwidth (4GE per node).
The compression card improves Hadoop performance by 20%, and reduces network traffic by more than 60%, CPU usage by over 80%, used disk space by more than 50%, and disk I/O data read and write operations by over 50%.

 

RISC to IA,Unix Migration

Huawei E9000 for RISC to IA Unix migration 

Typical Configuration

E9000 chassis
CH121 2-socket compute node (for web, middleware, and application VM deployment)
CH240 4-socket compute node (database physical machine deployment)
CH280 8-socket compute node (2013)
PCIe SSD card
CX110 and CX311 switch modules

Features

Supports the VM resource pool for web, application, and middleware layers and virtual NICs to increase network throughput, which ensures VM security.
The OLTP database is deployed on physical machines, which supports stateless computing and migration, and use FC SAN for shared storage.
Supports a large-capacity cache by using PCIe SSD cards to improve I/O performance for databases.
Supports highly reliable redundancy for PSMs (N+1/N+N) and fan modules.

High Performance Computing (HPC) Application

Huawei E9000 HPC application

Features

CH140 Twin Computing Node: High-density computing
Supports 2*2-socket compute nodes in half-width slots
Supports 64 Intel Xeon E5-2600 processors in a chassis.
Supports 192 E5-2600 processors in a rack(3 chassises).
Supports 12 TB memory capacity
Supports InfiniBand QDR/FDR
Supports InfiniBand EDR evolution
Supports GPU and SSD resource expansion

 

 

Sample Configuration

Huawei E9000 base with 6 compute nodes and 10GE uplinks

No. Part Number Model Description Unit Qty. Qty. Note
  E9000 Basic Package: 6 CH121v3 10G Nodes with 16*10G Uplink Switch Module      
1 E9000   1  
1.1 Chassis Basic Configuration      
  02301029 IT1K21E9000 12U standard integrative module with  Delivery Accessory 1 1  
  03054675 IT1DSMMA0000 MM910,Shelf Management Module A 2 2  
  03030QEJ IT0E01FAN000 OSCA Fan Box 14 14  
  02310LKL IT1MPOWERM00 3000W platinum AC power supply unit 6 6  
1.2 Switch Module      
  03054849 IT1DXCUB00 CX310,16*10GE Converged Switch Module 2 2 16 x 10GE uplink to switch OUTSIDE E9000 Chassis; 32 x 10GE downlink to compute nodes within E9000; e.g: CH121v3 in this case.
1.3 Optional Equipments      
  21240817 EGUIDER05 4U Static Rail Kit 1 1  
             
  CH121 V3_Site1      
2 E9000   6  
2.1 Blade Basic Configuration      
2.1.1 Blade MotherBoard      
  03056132 IT11SGCA05 Grantley EP Compute Node,CH121 V3 1 6  
2.1.2 Haswell EP CPU      
  41020498 EHSE52620 Intel Xeon E5-2620 v3(2.4GHz/6-core/15MB/85W) Processor 2 12  
2.1.3 DDR4 Memory      
  06200201 N21DDR432 DDR4 RDIMM Memory,32GB,2133MT/s,2Rank(2G*4bit),1.2V,ECC 4 24  
2.1.4 Hard Disk(include Front Panel)      
  02311HAL N900S1210W2 HDD,900GB,SAS 12Gb/s,10K rpm,128MB or above,2.5inch(2.5inch Drive Bay) 2 12  
2.1.5 RAID Card and Other Accessories      
  03021FTX BC0MESMCE600 RU120 SAS/SATA RAID Card,RAID0,1,10,1E,0 Cache(LSI2308) 1 6  
2.1.6 Network Mezz Card      
  03022YMY IT11MXEM MZ310,2*10G Mezzanine Card 1 6 Network Adapter On Compute Node; 2 Port 10G connect to switch module internally, e.g: CX310 in this case

 

Huawei E9000 base with 8 compute nodes, 10GE uplink and 8G Fibre Channel connections

No. Part Number Model Description Unit Qty. Qty. Note
  E9000 6 Compute Node with 10G Ethernet uplink and 8G SAN Package      
1 E9000   1  
1.1 Chassis Basic Configuration      
  02301029 IT1K21E9000 12U standard integrative module with  Delivery Accessory 1 1  
  03054675 IT1DSMMA0000 MM910,Shelf Management Module A 2 2  
  03030QEJ IT0E01FAN000 OSCA Fan Box 14 14  
  02310LKL IT1MPOWERM00 3000W platinum AC power supply unit 6 6  
1.2 Switch Module      
  03056858 IT11XCUA08 CX912,16*10GE Port,8*8G FC Port(4-Uplink Port Activate,Brocade),Multi-Plane Switch Module 2 2 16 x 10GE uplink to switch OUTSIDE E9000 Chassis; 32 x 10GE downlink to compute nodes within E9000; e.g: CH121v3 in this case.
8*8G FC Port connects to
OUTSIDE SAN Storage;can be extended to 16*8G FC with license.
  05330492 BC0LICENSE02 FCSwitch-Brocade-ISL Trunking License License 2 2  
  05330493 BC0LICENSE01 FCSwitch-Brocade-Full Fabric License 2 2  
1.3 Optional Equipments      
  21240817 EGUIDER05 4U Static Rail Kit 1 1  
             
  CH121 V3_Site1      
2 E9000   6  
2.1 Blade Basic Configuration      
2.1.1 Blade MotherBoard      
  03056132 IT11SGCA05 Grantley EP Compute Node,CH121 V3 1 6  
2.1.2 Haswell EP CPU      
  41020498 EHSE52620 Intel Xeon E5-2620 v3(2.4GHz/6-core/15MB/85W) Processor 2 12  
2.1.3 DDR4 Memory      
  06200201 N21DDR432 DDR4 RDIMM Memory,32GB,2133MT/s,2Rank(2G*4bit),1.2V,ECC 4 24  
2.1.4 Hard Disk(include Front Panel)      
  02311HAL N900S1210W2 HDD,900GB,SAS 12Gb/s,10K rpm,128MB or above,2.5inch(2.5inch Drive Bay) 2 12  
2.1.5 RAID Card and Other Accessories      
  03021FTX BC0MESMCE600 RU120 SAS/SATA RAID Card,RAID0,1,10,1E,0 Cache(LSI2308) 1 6  
2.1.6 Network Mezz Card      
  03022XBT IT11MXEL MZ912-2*10GE,2*8G FC Mezzanine Card 1 6 Network/SAN Adapter On Compute Node; 2 Port 10G connect to switch module internally, e.g: CX912 in this case
2*8G FC Connects
internal SAN switch, in this case, CX912

  

Blade server price HP C7000 Dell M1000e Huawei E9000

Huawei E9000 Blade Server

Installed in a standard 19-inch rack, the E9000 is 12 U high. Its dimensions are: 530.2 mm x 442 mm x 840 mm (H x W x D)
An E9000 chassis houses 16 half-width slots or 8 full-width slots
Management modules support 1+1 redundancy and management protocols and interfaces such as IPMI 2.0, SOL, SSL, SSH, and web
Supports up to 6 Platinum PSMs (AC 3000 W/DC 2500 W) in N+N/N+1 redundancy with up to 95% efficiency under 50% load and supports dynamic management for PSMs in the hibernation state
Provides 14 fan modules(3 groups) and  every group supports single-fan failures

For more information on E9000 Blade server, please check the product documentation here.

 

Competitive products in the market:

HPE C7000, C3000 , Superdome X

Dell PowerEdge M1000E

SuperMicro SuperBlade GPU/Xeon Phi B;ade TwinBlade NVMe StprageBlade NVMe DatacenterBlade

Lenovo/IBM Flex System, BladeCenter S. BladeCenter H, BladeCenter HT

 

Quote on Huawei E9000 chassis and blades

Quote on Huawei E9000 v3 chassis and blades

Quote on Huawei E9000 typical configuration

Contact us:Tel: 1-866-318-6636 (US/CANADA)

     Email: info@actfornet.com