By Jeremy Pedersen
Welcome back for the sixth installment in our weekly Q&A blog series! We're taking a look back into the archives this week, and addressing some questions from previous Alibaba Cloud training sessions.
By default ECS instances include a single Cloud Disk, which serves as the system disk (boot disk) for the instance.
The default size for the system disk is 40 GB for most instance types, and for newer instances (generation 6 and up) the disk type is Enhanced SSD (ESSD).
As I'm writing this in April 2021, we have 23 Regions and 69 Zones, but that number is growing all the time!
You can see a nice list (and map) of our locations here.
If you're a developer and want more details (like region endpoint names), look here, in the ECS documentation.
Yes, absolutely. The procedure is different depending on whether the disk is a data disk (non-boot disk) or system disk (boot disk) but in both cases it's a straightforward process. Have a look at this page from the ECS documentation.
Disk backups (called snapshots) are Zone-independent by default, and can be used in any Zone within a region. So yes, you can clone disks across Zones. The process is:
Not necessarily, actually. The Alibaba Cloud snapshot system is smart enough to preserve the state of a chain of snapshots even if intermediate snapshots are deleted. It does this by breaking each snapshot into a series of "blocks", making sure to only throw away blocks from a deleted snapshot if they are not referenced by other snapshots.
We have a nice diagram illustrating this here in the ECS documentation.
Let's have a quick example:
I then delete #2. What will happen? The snapshot system will compare snapshots #2 and #3 to see which blocks from 2 are referenced by #3. It will then keep those blocks which are still needed by #3 (essentially, these become part of snapshot #3).
On 2021-04-30, when I create my next snapshot, it will be an incremental snapshot based on the changes to the cloud disk made since the last snapshot, #3, so I don't need to create a new full snapshot!
Yes, Security Groups are stateful: they work more or less like a traditional firewall in that respect.
Yes, automatic snapshot policies can both automatically create and delete snapshots. In addition to specifying when snapshots should be taken, the automatic snapshot policy also sets the retention period for snapshots. Snapshots are automatically deleted after this retention period passes.
We don't have an explicit proximity placement feature, but placing instances within the same some will ensure that they end up within the came Alibaba Cloud datacenter.
If you want to ensure your ECS virtual machines all get placed on the same physical host for maximum speed (at the cost of disaster tolerance), you can do that by launching your ECS instances on a Dedicated Host.
If you are trying to make sure instances are in the same Zone but spread out across different physical hosts, for a mix of low latency and high availability, you can try out Deployment Sets.
Yes, we have what we call local disks. These are storage that is physically attached to a given ECS instance, providing very high performance I/O.
No: you can launch load balancers that do not have a public IP attached.
Currently there are two types of Alibaba Cloud load balancer:
In the case of the shared performance Classic Load Balancer, it's even possible to create an instance that is free, assuming it:
So that's a good option for non-internet (internal) load balancing use-cases!
Yes, OSS buckets provide private endpoints that are accessible to any ECS instance that is in a VPC in the same region as the OSS bucket.
So, if I have an OSS bucket in Shanghai and I want to access it over Alibaba Cloud's private network, I can use the OSS internal endpoint for Shanghai.
You can find the endpoint address for each of your buckets on the Overview page in the console.
You can see a global list of OSS endpoints (internal and external) here.
Great! Reach out to me at jierui.pjr@alibaba-inc.com
and I'll do my best to answer in a future Friday Q&A blog.
You can also follow the Alibaba Cloud Academy LinkedIn Page. We'll re-post these blogs there each Friday.
Friday Q&A - Week 5 - Cost Control On Alibaba Cloud: Getting More For Your Dollar (Part 2)
JDP - March 26, 2021
JDP - May 7, 2021
JDP - April 9, 2021
JDP - June 4, 2021
JDP - May 20, 2021
JDP - July 9, 2021
Respond to sudden traffic spikes and minimize response time with Server Load Balancer
Learn MoreIndustry-standard hardware security modules (HSMs) deployed on Alibaba Cloud.
Learn MoreTair is a Redis-compatible in-memory database service that provides a variety of data structures and enterprise-level capabilities.
Learn MoreElastic and secure virtual cloud servers to cater all your cloud hosting needs.
Learn MoreMore Posts by JDP