AWS 上用空間買 IOPS 的故事...

在「A web performance issue」這邊講到 Mozilla 的系統產生效能問題,後續的 trouble shooting 以及解決問題的方案。

這個系統跑在 AWS 上,在一連串確認後發現是 RDS 所使用的 EBS 的 IOPS 滿了:

After reading a lot of documentation about Amazon’s RDS set-up I determined that slow downs in the database were related to IOPS spikes. Amazon gives you 3 IOPS per Gb and with a storage of 1 Terabyte we had 3,000 IOPS as our baseline. The graph below shows that at times we would get above that max baseline.

然後大家對於解法都差不多,因為 Provisioned IOPS 太貴,所以直接加大空間換 IOPS 出來 (因為 General SSD 裡 1 GB 給 3 IOPS):

To increase the IOPS baseline we could either increase the storage size or switch from General SSD to Provisioned IOPS storage. The cost of the different storage type was much higher so we decided to double our storage, thus, doubling our IOPS baseline. You can see in the graph below that we’re constantly above our previous baseline. This change helped Treeherder’s performance a lot.


In order to prevent getting into such a state in the future, I also created a CloudWatch alert. We would get alerted if the combined IOPS is greater than 5,700 IOPS for 6 datapoints within 10 minutes.

不過 General SSD 的 IOPS 是沒有 100% 保證的,只有這樣寫:

AWS designs gp2 volumes to deliver 90% of the provisioned performance 99% of the time.


AWS 的 EBS 預設型態改為 GP2 (SSD)

AWS 宣佈 EBS 的預設型態從 Standard 變成 GP2:「EBS default volume type updated to GP2」。

包括 web console 與 API 的預設值都改成 GP2:

The AWS console defaults to GP2 in all regions. On July 29th the default EBS volume type was updated in thirteen regions from Standard to GP2. Now AWS API calls for volume, image, and instance creation also default to GP2 in all regions.

GP2 是 SSD,所以可以提供比較低的 latency,而另外一個用 GP2 的好處是 i/o 的費用已經含在內了 (Standard 會另外收取費用),對於成本估算會比較簡單一些,尤其是 i/o 量比較大的時候。

AWS Cloud 的用法

Hacker News Daily 上看到這則,分享了 AWS (他的前東家,超過八年) 的使用經驗:

除了可以在 Twitter 上看以外,也可以用 Thread reader 直接讀整條 thread,應該也還算清楚:「This is how I use the good parts of @awscloud, while filtering out all the distracting hype.」。

這邊的經驗談主要是在 web 與 app 相關的服務這塊:

有講到 AWS 的業務其實圍繞在 scalability 上發展,但這對 startup 可能反而是扣分,因為暴力法解反而可以大幅簡化架構換得 agile (而讓 startup 存活下來)。

另外從團隊的開發成本來看,這些 scale 的技術增加了開發成本,產生了很多開發上的限制,這些觀點也有點帶到「Premature optimization is the root of all evil」在講的事情:


除了 DynamoDB 的意見不同外 (這邊提到的 DDB),其他的我都可以接受...

Amazon EBS 的預設加密機制

EBS 有選項可以預設開加密了:「New – Opt-in to Default Encryption for New EBS Volumes」。


Per-Region – As noted above, you can opt-in to default encryption on a region-by-region basis.


AMI Sharing – As I noted above, we recently gave you the ability to share encrypted AMIs with other AWS accounts. However, you cannot share them publicly, and you should use a separate account to create community AMIs, Marketplace AMIs, and public snapshots. To learn more, read How to Share Encrypted AMIs Across Accounts to Launch Encrypted EC2 Instances.

然後有些服務有自己的 EBS 設定,這次不受影響。而有些底層其實是用 EC2 的服務 (然後開 EBS) 就會直接套用了:

Other AWS Services – AWS services such as Amazon Relational Database Service (RDS) and Amazon WorkSpaces that use EBS for storage perform their own encryption and key management and are not affected by this launch. Services such as Amazon EMR that create volumes within your account will automatically respect the encryption setting, and will use encrypted volumes if the always-encrypt feature is enabled.

AWS 給 EBS 用的 Data Lifecycle Manager 在東京可以用了?

先前在「Amazon EBS Snapshot 支援 Lifecycle Management」這邊提到 AWS 設計了 Data Lifecycle Manager,讓 EBS 磁碟可以自動產生 snapshot 並且管理保留份數,可以當作某種備份機制。


Availability – Data Lifecycle Manager is available in the US East (N. Virginia), US West (Oregon), and Europe (Ireland) Regions.

剛剛發現在東京也已經可以用了?但好像沒看到有公告提過... 設下去看看會不會動好了。

Amazon EBS (gp2) 提昇效能...

AWS 宣佈提昇了 Amazon EBS (gp2) 的效能:「Amazon EBS Increases Performance of General Purpose SSD (gp2) Volumes」。

本來上限是 10k IOPS,現在提升到 16k IOPS 了。另外最大傳輸速度也從 160 MB/sec 提升到 250 MB/sec:

Today we are announcing a 60% improvement in performance of General Purpose SSD (gp2) Volumes from 10,000 IOPS to 16,000 IOPS and from 160 MB/s to 250 MB/s of throughput per volume.

應該還是維持 3 IOPS per GB 的設計,但這對於想要用 gp2 堆效能的人來說算是好消息,可以用更大的空間堆出更多 IOPS 了... (像是「Percona 的人接受 AWS 的建議,重新測試了 Percona XtraDB Cluster 在 gp2 上的效能...」這篇)

EC2 支援休眠

AWSEC2 instance 推出了休眠模式:「New – Hibernate Your EC2 Instances」。

休眠模式主要是透過保留記憶體內的內容 (會寫到 EBS 裡),讓機器開起來比較快,不過這邊有要求 root EBS volume 需要加密:

When an instance is instructed to hibernate, it writes the in-memory state to a file in the root EBS volume and then (in effect) shuts itself down. The AMI used to launch the instance must be encrypted, as must the root EBS volume of the instance. The encryption ensures proper protection for sensitive data when it is copied from memory to the EBS volume.

費用的部分是 EBS volume 與 Elastic IP:

While the instance is in hibernation, you pay only for the EBS volumes and Elastic IP Addresses attached to it; there are no other hourly charges (just like any other stopped instance).

一般對外服務的伺服器好像用不太到... 應該是內部系統或是開發機的需求?

Amazon EBS Snapshot 支援 Lifecycle Management

以往用 Amazon EBS Snapshot 需要用 Lambda 當 cron job 建立 snapshot,以及管理要存的數量 (要刪掉舊的),現在 AWS 直接提供服務幫你處理:「New – Lifecycle Management for Amazon EBS Snapshots」。文章的截圖就說明了這個新功能:


You can create and use Data Lifecyle Manager policies at no charge; you pay the usual storage charges for the EBS snapshots that it creates.

Data Lifecycle Manager is available in the US East (N. Virginia), US West (Oregon), and EU (Ireland) Regions.

先繼續用 Lambda,等東京有的時候再換過去...

EBS 可以動態擴充 Magnetic Volume 的大小了...

AWS 宣佈可以動態擴充 EBS (Magnetic Volume) 的大小了:「Amazon EBS Extends Elastic Volumes to Support EBS Magnetic (Standard) Volume Type」。EBS (Magnetic Volume) 被歸類到前個世代的產品,意外的把這個功能支援了前個世代的產品...

前世代的 Magnetic Volume 與 Throughput Optimized HDD (st1) & Cold HDD (sc1) 來比較起來貴了一些,但可以當作開機空間,另外低消只有 1GB (而 st1sc1 是 500GB),不過如果與 General Purpose SSD (gp2) 比的話,只要 I/O 有點量,整體的價錢就會超過 gp2,目前看起來還是用 gp2 會好一些...

Percona 的人接受 AWS 的建議,重新測試了 Percona XtraDB Cluster 在 gp2 上的效能...

去年年底的時候 Percona 的人在 AWS 上測試 Percona XtraDB Cluster 的效能,尤其是針對底層應該選擇哪種 EBS 的部分給了一些建議。可以參考先前寫的「Percona 分析在 AWS 上跑 Percona XtraDB Cluster 的效能 (I/O bound)」這篇。

當時的建議是用 io1,雖然是比較貴,但對於效能比較好。

而後來 Percona 的人收到 AWS 工程師的建議,可以用另外一個方式,可以在 gp2 上拉出類似的效能,但成本會比 io1 低不少:「Percona XtraDB Cluster on Amazon GP2 Volumes」。

這個方式是利用 gp2 會依照空間大小,計算可用的 IOPS。在官方的文件裡是這樣描述 gp2 的效能 (IOPS):

General Purpose SSD (gp2) volumes offer cost-effective storage that is ideal for a broad range of workloads. These volumes deliver single-digit millisecond latencies and the ability to burst to 3,000 IOPS for extended periods of time. Between a minimum of 100 IOPS (at 33.33 GiB and below) and a maximum of 10,000 IOPS (at 3,334 GiB and above), baseline performance scales linearly at 3 IOPS per GiB of volume size. AWS designs gp2 volumes to deliver the provisioned performance 99% of the time. A gp2 volume can range in size from 1 GiB to 16 TiB.

在這個前提下,需要 10000 IOPS 的效能會需要 3.3TB 以上的空間,所以 Percona 就被 AWS 的工程師建議直接拉高空間重新測試:

After publishing our material, Amazon engineers pointed that we should try GP2 volumes with the size allocated to provide 10000 IOPS. If we allocated volumes with size 3.3 TiB or more, we should achieve 10000 IOPS.


接下來就比較儲存成本,大約是 io1 版本的一半價錢:

如上面文件中提到的,gp1 不完全保證效能,但統計出來經常能夠提供出 3 IOPS/GB 的效能。而 io1 則是保證效能,不太需要擔心效能不穩定的問題。就是這個差異,反應到成本上面就有蠻大的差距。善用這點設計系統,應該會對整體成本有蠻大的幫助... (但對 latency 就未必了,尤其是 P99 之類的數值)