Skip to end of metadata
Go to start of metadata

Introduction

Data is provided for information purposes only. Usage may vary for a same number of users, depending on hardware structure and use habits. Many factors can affect usage: email volume, email size, number of recipients, number of events, event scheduling, etc.

On this page:

Related:

About units

Several BlueMind components use up resources. A typical "per user" calculation cannot be applied because a user who only uses messaging will not generate the same load for the system as a user using messaging and collaborative services (Calendar, etc.) and a PDA.

As a result, sizing is calculated "per unit", on the following basis:

User profile

Units

Messaging only

1

Messaging + intense collaborative use

2

Messaging + collaborative services + PDA

5

Also, for a same amount of units, a use of messaging only won't consume the same amount of resources as a messaging+collaborative use: unlike collaborative tools, messaging, for example, is more heavily dependent on IO than on CPU.

CPU

CPU is stated in number of cores. Reference values are based on recent Xeon-type CPU.

BlueMind has several services, as a result we recommend a minimum of 2 cores.

Units

Number of core(s)

1-50

2

50-1000

4

1000+

1 / 250 units

 

RAM

Units

RAM

1-50

6 GB

50-500

12 GB

500+

12 + 1 GB / 500 units  

Storage / IO Performance

IOPS = "Input/Output Operations Per Second"

The messaging service is a heavy user of IO, as a result storage is sized in IOPS . As for storage space, it depends on client requirements (quotas, etc.)

Units

IOPS per unit

1

.3  

IOPS data for storage devices

Device  

Type  

IOPS  

Interface  

Notes

7,200 rpm SATA drives

HDD

~75-100 IOPS

SATA 3 Gb/s 


10,000 rpm SATA drives

HDD

~125-150 IOPS

SATA 3 Gb/s

 

10,000 rpm SAS drives

HDD

~140 IOPS

SAS

 

15,000 rpm SAS drives

HDD

~175-210 IOPS

SAS


   

Source: http://en.wikipedia.org/wiki/IOPS#Examples

Examples

Core/RAM distribution over several servers (virtual or otherwise) is not described here.

However, for up to 16/24 cores, we believe that a single-platform installation makes sense.

Above this threshold, and to manage populations of tens of thousands of users or more, the architecture must be distributed.

Also, the messaging part as well as the database (which collaborative use/PDA places heavy demands on) must be kept separate from the rest.

Users / Units

CPU

#cores

RAM

IOPS / Disk

25 users / 5 with PDAs

45 units (20 + 25)

2

6

13.5 / all disks

150 users / 50 collaborative users of which 25 with PDAs

225 units (100+25*2+25*5)

4

12

67.5

SATA 7,200 minimum

300 users / 100 collaborative users / 30 PDAs

490 units (200 + 70*2 + 30*5)

4

12

147

2 * 10K rpm SAS

1 * 15K rpm SAS

600 users / 200 collaborative users / 50 PDAs

950 units (400 + 150*2 * 50*5)

4

14

285

SSD, Bay or other system

1,000 users / 250 collaborative users / 100 PDAs

1,300 units (750 + 150 * 2 + 100 * 5)

6

15

390

SSD, Bay or other system

2,000 users / 500 collaborative users / 200 PDAs

3,100 units (1500 + 300*2 + 200 * 5)

13

16

930

Bay (2,000 IOPS)

4,000 users / 1000 collaborative users / 300 PDAs

5,900 units (3000 + 700*2 + 300*5)

24

22

1,770

Bay (2-3,000 IOPS)

4,000 users / 1000 collaborative users / 1000 PDAs

8,000 units (3000 + 1000*5)

32

26

2,400

Bay 3,000 IOPS

SAN / other technology

4,000 users / 4,000 collaborative users / 1000 PDAs

1,1000 units (3,000*2 + 1,000*5)

44

32

3,300

SAN / other technology

5,000+ users (10,000; 100,000; etc.)

The system must be distributed and the architecture designed on an ad-hoc basis.

   

Bandwidth

Bandwidth requirements cannot be predicted as they largely depend on mail traffic.Please note that the data on bandwidth usage of the BlueMind calendar and PDAs below clearly shows the preponderance of mail traffic.

BlueMind Calendar bandwidth

For a user with the Calendar application open in their web browser, in http and in bytes (measured on the network with Wireshark):

  • every 30 seconds: one doSync 1067 / 293 (sends local modifications and retrieves changes)
  • every 5 seconds: one ping: 898 / 233, i.e. 5388 / 1398 in 30s (one keepalive)

Client to server: 215 bytes/sec (1,067+5,388)/30

Server to client: 56 bytes/sec (293+1,398)/30

Number of active users

Client to Server

Server to Client

1

215 B/s

56 B/s

100

21 KB/s

6 KB/s

1,000

210 KB/s

60 KB/s

10,000

2.1 MB/s

600 KB/s

With room for maneuver, for 1,000 Calendars running in web browsers:

  • Client to server: 500KB/s
  • Server to client: 150KB/s

Contacts bandwidth

For a user with the Contacts application running in their browser, in http and in bytes:

144 bytes/second

Specifically:

  • a ping every 5 seconds
  • a "bmc" every 30 seconds

By doubling the value measured to ensure a comfortable safety margin, we calculate a bandwidth of 288 bytes per second for a user who has launched the Contacts application.

PDA bandwidth

Microsoft provides the following ActiveSync ratios: 1.04KB/s/user

i.e. for 100 PDAs: 104Kbit, or 13KB/s

By taking a sensible safety margin of x2, we calculate:

  • 100 PDAs == 26KB/s
  • 1,000 PDAs == 260KB/s

Enregistrer

Enregistrer

Enregistrer

  • No labels