SaaS-type cloud

What is SaaS?

전통적인 소프트웨어 판매 센터는 사용자에게 소프트웨어 패키지 제품의 라이센스를 판매하는 형태로, 사용자는 자신이 가진 컴퓨터에서 소프트웨어를 실행하고 이용하는 형태였습니다.
SaaS란 소프트웨어 공급자측의 컴퓨터에서 소프트웨어를 실행시켜 사용자는 인터넷 등의 네트워크를 통해 서비스를 이용하고 그에 대한 서비스 비용을 지불하는 형태입니다.
SaaS(Software as a Service)는 소프트웨어 및 관련 데이터는 중앙에 호스팅되고 사용자는 웹 브라우저 등의 클라이언트를 통해 접속하는 형태의 소프트웨어 전달 모델입니다.
기존의 ASP(Application-Service-Providers)와 동일하며 최근에는 클라우드 컴퓨팅으로도 불립니다.
네트워크 기반으로 접근하고 관리하는 상업적으로 사용 가능한 소프트웨어로 각 고객 사이트가 아닌 중앙의 위치에서 관리, 고객이 웹 모바일 등을 통해 애플리케이션에 접근하도록 하며 애플리케이션 전달은 일반적으로 일대일 모델과 일대다 모델이 있습니다.
Eliminating the need to download patches and upgrades by centralized update function.


What is SaaS-type on Shared Server and Single Server?

SaaS-type on Shared Server
(Public Cloud)
  • Service Providers provide service and management
  • Many companies are set up in one server to use
  • Low Price
  • May be affected by other companies
SaaS-type on Single Server
(Private Cloud)
  • Service Providers provide servers and management
  • Only one company is exclusively set on one server
  • High reliability and security
  • High price
  • Exclusive use of a single physical server

About Server management capabilities of Hanbiro Inc.

Hanbiro Inc. has a few dozen Linux engineers and is operating more than 5,000 servers.
We offer customers all server operations such as building hardware of server, OS management, Application management, etc.
We got know-how to minimize downtime by various automated software deployment and operation of the server operating since 1999.
We are operating customers' server in the best conditions with automation of server operations and fault notification process manuals.
More than 1,000 mail servers are operating.


Stability of the groupware server operating information and data

The default setting of SaaS in Public Server

By setting a DRBD to minimize data loss

DRBD (Distributed Replicated Block Device) is a distributed replicated storage system for the Linux platform.
DRBD makes it possible to maintain consistency of data among multiple systems in a network.
The advantage is that it is not affected even if the data is increased because will transfer data at the block level by techniques of real-time data synchronization between A server and B server through network.
DRBD's synchronization algorithm is efficient in the sense that only those blocks that were changed during the outage must be resynchronized, rather than the device in its entirety.
일반적으로 사용하는 데이터 동기화 기법에서 나타나는 저장 데이터의 크기 증가 또는 파일 개수가 많아지면 동기화에 따른 부하 및 시간이 많이 소요되는 단점을 극복하기 위한 방식입니다.
네트워크에 연결된 두 대의 서버 내 데이터를 동시에 동기화 하는 기술로 파일 단위가 아닌 블록 단위로 데이터를 이동하기에 데이터의 증가에 따른 영향을 덜받는 장점이 있습니다.
It is implemented as a kernel module management applications and some shell scripts and is normally used on high availability (HA) computer clusters.
DRBD can be understood as network based RAID-1.
DRBD also refers to the logical block devices provided by the scheme and to the software that implements it.

The advantage of composition
  • The reliability of the data maintained by implementing redundancy though data on storage server is replicated in real-time.
    Can be restored immediately when failure of the storage server occurs
  • Hanbiro can cope with failures quickly thanks to many years of experience operating server
  • Handling failure situations and downtime
    Because the configured redundant storage servers will take place the role automatically so downtime can be minimized
    Failure time that can occur: less than 5 seconds
Operational know-how

Can be aware of the contents of a failure to respond to each server through real-time monitoring
Even in case you add a server based on service volume increases, additional support after the same configuration with the original running servers
Available for prompt action in unexpected situations thanks to technology experienced long-term service proceeding

Data Backup Schedule

Data Backup Schedule
  Periodic data backup Disk image backup
Cycle Backup every morning Backup every morning
Method Incremental backup Incremental backup
Store 30 days 2 days

Service interruptions and data loss due to hardware failure

Service interruptions and data loss due to hardware failure
Hardware Parts Setting Information Number of failed servers Recovery time Data loss
Server DRBD 2 When one server failed, service disruption is not applied.
When two servers failed, recovery within two hours
None
Hard disk installed OS SoftRaid 1 2 When one hard disk failed, service disruption is not applied.
When two hard disks failed, recovery within two hours
None
Data stored in the hard disk Raid 1+0 2 When one hard disk failed, service disruption is not applied.
When two hard disk failed, emergency repair within two hours
Secondary recovery within a day
Data loss from last backup until disk failure
Motherboard   1 Recovery within two hours None
Power Supply   2 When one Power Supply failed, service disruption is not applied.
When two Power Supplies failed, recovery within two hours
None
Network     Can not predict based on IDC or ISP None
Supervisor
Moon Sori
1544-4755 Ext. 505
sori@hanbiro.com
(English Consultation Available)