What's the maximum database size for WikidPad?
149 views
Skip to first unread message
Forrest
Mar 14, 2009, 12:06:31 PM3/14/09
to wikidpad
I'm an Evernote user and realized it becomes slow when the database
getting pretty big, say 500M. I'm wondering if WikidPad has this
issue. Any reply is appreciated.
getting pretty big, say 500M. I'm wondering if WikidPad has this
issue. Any reply is appreciated.
Lee Cook
Mar 14, 2009, 12:58:00 PM3/14/09
to wiki...@googlegroups.com
Forrest
Mar 14, 2009, 8:14:04 PM3/14/09
to wikidpad
Thanks LeeJai. Could I ask how big your WikidPad databasee is?
On 3月15日, 上午12时58分, Lee Cook <leejaic...@yahoo.com> wrote:
> Saturday March 14 2009. Ms. Lee Jai Cook, Roaring Springs, Texas USA.
>
> Hello, Forrest: I am not an Evernote user.
> However, I wish to say that when I search for an article in Wikipedia Encyclopedia the response is virtually instantaneous.
> Best wishes, LeeJai
>
> ________________________________
> From: Forrest <hiForr...@gmail.com>
On 3月15日, 上午12时58分, Lee Cook <leejaic...@yahoo.com> wrote:
> Saturday March 14 2009. Ms. Lee Jai Cook, Roaring Springs, Texas USA.
>
> Hello, Forrest: I am not an Evernote user.
> However, I wish to say that when I search for an article in Wikipedia Encyclopedia the response is virtually instantaneous.
> Best wishes, LeeJai
>
> ________________________________
.cn zhoujb
Mar 14, 2009, 9:39:37 PM3/14/09
to wiki...@googlegroups.com
Hi, Forrest
As the backend of Wikidpad is SQLite, I think the following message
may be helpful.
===============================================
While SQLite’s databases can scale to 2 terabytes, there are memory
(RAM) costs associated with large databases. When SQLite begins a transaction, it allocates
a bitmap for tracking dirty pages, which assists in managing its rollback journal. To
do this, it requires 256 bytes of RAM for every 1MB of database. Thus, when databases
become very large, the size of the bitmap allocated in each transaction can become
significant. A 100GB database would require 25MB of RAM to be allocated before each
transaction. This directly affects the rate at which transactions can be performed regardless
of their complexity. So in reality, the practical limits on database size are in the tens of
gigabytes. They can be much bigger, but keep in mind that the overhead associated with
transaction startup will increase linearly with the database size.
===============================================
zhou
As the backend of Wikidpad is SQLite, I think the following message
may be helpful.
===============================================
While SQLite’s databases can scale to 2 terabytes, there are memory
(RAM) costs associated with large databases. When SQLite begins a transaction, it allocates
a bitmap for tracking dirty pages, which assists in managing its rollback journal. To
do this, it requires 256 bytes of RAM for every 1MB of database. Thus, when databases
become very large, the size of the bitmap allocated in each transaction can become
significant. A 100GB database would require 25MB of RAM to be allocated before each
transaction. This directly affects the rate at which transactions can be performed regardless
of their complexity. So in reality, the practical limits on database size are in the tens of
gigabytes. They can be much bigger, but keep in mind that the overhead associated with
transaction startup will increase linearly with the database size.
===============================================
zhou
Message has been deleted
Forrest
Mar 14, 2009, 10:57:59 PM3/14/09
to wikidpad
Thank you, Zhou. What you've provided is very helpful and I decide to
have a go for WikidPad.
have a go for WikidPad.
Lee Cook
Mar 15, 2009, 1:03:20 PM3/15/09
to wiki...@googlegroups.com
Reply all
Reply to author
Forward
0 new messages