VMWare vSAN Series – Architecture

Hi everybody,

in reference to my prior article vSAN – High Available Solution this article will explain the basic architecture and give you some insight into the technical details.

vSAN Basic Architecture

There are several different approaches to vSAN cluster. The decision has to be made stretched/non stretched and direct connect/multi node based on the requirements such as availability/performance/scaleability

Stretched: at least two nodes are stretched over two sites; additional requirements – 10Gbit and <5ms rtt (round trip time) – layer 3 supported; witness required

Non Stretched: all nodes are on the same site

Direct Connect: just two nodes without redundant ethernet switches, witness required

Multi node: (redundant) 10 Gbit infrastructure required; up to 64 nodes in one cluster

Irrespective which deployment type you choose, every node has in common that it contributes local CPU / RAM / Network / Storage resources to the attached cluster.

It is possible to add nodes to the cluster without local storage, which just offer computing.

Network Storage Backend

Rule of thumb is that in hybrid configuration 1 Gbit is supported but not recommended.  Opposite is all-flash where 10 Gbit is mandatory. If you use PCIe NVMe card which provide transfer rates over 3500 MB/s you should consider 40 Gbit or even 100 Gbit. Which is at least cheap in 2 node direct connect configurations because you wont need that expensive switch infrastructure.

To deliver the goods we need at least one more VMKernel for the vSAN traffic.

If you need to use a Witness you need to tag you management VMKernel with Witness traffic. So far this is just possible over the command line:

esxcli vsan network ip add -i vmk0 -T=witness

Internal Disk Groups

Every disc group consists of exactly one flash caching device (SAS,SATA or NVME) for read caching and write buffering operations – Caching Tier

… and up to 7 flash or spinning disks which offer storage capacity – Storage Tier

Important is that in all-flash configurations the caching tier is 100% write buffer.

In hybrid configurations (capacity tier is HDD) the caching tier is 30% write buffer and 70% read cache.

 

After creation its formatted in VMFS-L and consequently every Disk group is now a silo for vSAN objects.

In addition the max. amount of disk groups per host is five.

vSAN Datastore and Storage Policy

In order to create a vSAN Datastore you need existing disk groups which will be aggregated together and compose the vSAN Datastore.

Therein objects can be saved fulfilling the requirements which will be mapped in the storage policy.

 

So every virtual machine is attached to a storage policy which contains some basic and optional advanced configuration parameters for storage performance and redundancy.

There a two key parameters which are mandatory in the storage policy.

Stripe width (SW) – object placement striped on different physical disks in order to adapt to performance requirements. Objects sized above 255 GByte are divided automatically into more stripes.

Faults to tolerate (FTT) – compensation of failing nodes, fault to tolerate = 1 means all object information are at least saved into two different disk groups on different hosts. In conclusion you get availability at the cost of storage when you increase this parameter

Additional rules can be defined like “flash read cache reservation” or “IOPS limit for object”

Example vSAN Default Storage Policy:

 

Best regards

Nicolas Frey

VCAP6-DCV Design its a pass

Hi there,

after passing the VCAP Deploy exam, I successfully passed the VCAP Design exam. Thereby I am now VSphere Certified Implementation Expert Data Center Virtualization 6. (VCIX6-DCV)

Furthermore I tinker with the idea of completing the data center track. Which requires a vast expenditure both monetary and time consuming. The requirement is to successfully submit and defend an original design in front of a panel of experts. In Europe this is just possible in Staines, UK.

This process is far more complex than any other exam that exists in the IT universe. Currently there are just 264 VCDX certified people around the world, collectively over all tracks.

https://vcdx.vmware.com/

Along the way I prepare myself for the NCDA – Netapp Certified Data Administrator certification. Due to my participation at Insight 2017 in Berlin last week I know Netapp and Clustered Ontap will keep me occupied over the next years too. Really awesome data fabric features that will shape both private and public clouds.

BR, Nicolas Frey

VMWare vSAN Series – an overview of a hyper-converged – high available solution

Hello everybody,

got a new topic for you called VMWare vSAN which stands for virtual storage area network. This article will be an introduction in order to get a picture of the matter.

In the next part of this series I plan to write about all the features that come with it. Afterwards the architecture powered by an real life example will be explained. In the end I plan to really dive deep. A best practice series will follow which summarizes my experience and hopefully helps some desperate souls.

A different approach:

So to start, in the old days SAN was a very complex architecture which consists of storage arrays, network switches, fiber channel fabrics, (a redundant array of SAN switches) storage controllers, disk shelves and a lot of cabling. These silos had their own complex management and you needed a bunch of experts to implement the custom solution. In the end all that you got was some (high available) mass storage which you could access via different protocols like CIFS, NFS, iSCSI  and FC over the Ethernet or Fibrechannel over a dedicated storage network. Mostly no quality of service and very expensive. Furthermore it was the era of spinning disks which were power consuming, high latency and error-prone.

Besides the “old” dedicated storage approach another solution was invented. Nowadays you get not only storage but the whole stack. Computing, network and storage – on just a set of conventional servers with local (flash) disks. All components are abstracted and managed by a software. The term – software defined data center or in our case – software defined storage – emerged. This package offers a new idea of an economical, scale-able, low latency, high performance, compact, easy to administrate, maintainable, secure, feature heavy and future-proof solution.

This solution is called hyper-converged infrastructure and familiar software/hardware vendors, which I got a boundary-point with, are: DELL/VMWare, Nutanix and Simplivity.

As mentioned above I will start with the features of vSAN 6.6 in the next blog article so stay tuned.

Best wishes

Nicolas Frey

VMWorld Barcelona 2017 – DAY 1 and DAY 2

Hi folks,

arrived on Sunday via plane in Barcelona with about a hour delay due to high load at Barca airport. Check in at SB Plaza Europe around 11 pm. The location of the hotel is very nice. The Fira Grand Via (the fair where the VMWorld takes place) is in walking distance.

Next day I was very excited to take part at VMWorld for the first time. Got up very early and did a small workout in the fitness center at the hotel. Went for breakfast and again the coffee was the game changer. In addition the weather was awesome. Therefore I used the remaining time to go to the roof and shoot a picture that makes you jealous.

tldr? -> go for the pictures

Continue reading “VMWorld Barcelona 2017 – DAY 1 and DAY 2”

Mammutmarsch – Munich – August 2017 – 100km

Hi there,

got a new announcement for you.

Next month on Saturday the 05.08 I will attend a mind boggling 100km walk called Mammutmarsch. At some point i try to run a part of the distance. In any case the time limit is 24h. Registration is still open if you like to participate :).

Start and finish is in Munich and I will pass by two major lakes in Bavaria, Starnberger See and Ammersee. Challange accepted!

Furthermore the next defiance will be the Bodensee marathon on 16th September. Two days after the VMWorld, the excellent balance.

Pictures will follow.

Best regards

Nicolas Frey

 

 

Netapp Technology Forum Bavaria

 

Hello everbody,

last week on Thursday I attended a very nice event in the WWK – Arena in Augsburg. The host was Netapp and a lot of topics in form of presentations were covered. In addition I met a lot of sympathetic people.
I participated in various sessions and I listed the key facts of my favourites below:

 

Hardware and ONTAP 9.2

  • 32Gbit Fibrechannel SAN
  • 40 Gbit with 4x10Gbit break out connection
  • Hotplug Storage Controller – adding more performance – live
  • New ONTAP long term support release Modell -LTS (up to five years) and non LTS versions
  • NVE – Netapp volume encryption
  • Efficiency features
    • Aggregate (former volume) inline deduplication
    • Replication ( remote site needs more storage if inline data efficiency features are not supported )
    • Data efficiency features performance impact <10%
    • Opportunistic in nature – adaptive dedup
  • Fabric pool -> Capacity Tier -> Public or private cloud
  • QoS features in SAN
    • Min – max throughput per LUN
    • IOPs per LUN
    • Just in SAN

DevOps and Container

  • Concepts of app operation and software development
  • Standardize – consolidate – virtualize – automate – orchestrate
  • Digital innovation and transformation
  • DevOps:
    • Plan -> Develop -> Release -> Monitor -> Repeat
    • OSIMM
  • Non dynamic historic concepts replaced
  • For example: Apple release cycle – just a small part of users impacted
  • API SDK REST – Integration of common tools
  • Software defined infrastructure – environmental snapshots
  • DevOps as process of operation and development of infrastructures
  • New Job holders like “Site reliability engineer” (Google)
    • >50% Infrastructure
    • Predominant app developer
  • Container – differences to the traditional way (monolithic applications)
    • Micro services
    • Meshed over REST
    • IP based
    • Sufficient performance available nowadays
    • Can be upgraded independent
  • Concept of load balancing a.k.a horizontal scaling
  • For example: Google search engine
  • Containers are not virtual machines
  • Data storage out of the container
  • Container orchestration:
    • Kubernetes ‘k8s’
    • Dockerswarm
    • Mesos
  • Dynamic vs. persistent provisioning

GDPR – General data protection Regulation

 

  • GDPR called EU-DSGVO in Germany
  • Replaces the German “Datenschutzgesetz”
  • Affects every company which handles European personalised data or does Business with European companies
  • Will come into force on 25.05.2018
  • Save Harbour -> Privacy Shield -> GDPR
  • Company’s need additional personnel which are responsible – data protection Officer – privacy officer
  • Punishment for data leakage: up to 4% of yearly revenues or 20 million euros
  • Authorities must be informed within 72h
  • Everybody has the right to be forgotten
  • Data lineage
  • Physical location of the data must be documented
  • Data migration must be possible from one to another provider ( standardized file format, machine readable)
  • Concept of pseudonymization will be important
  • Encrypted data breach may not necessarily be reported to authorities
  • Code of conduct and certifications
  • Usage of hardening guides

Ontap and Cloud


Control of Hybrid Cloud

VMworld 2017 Barcelona – confirmed!

Hello everybody,

great news, this year I will be able to attend the VMworld for the first time, held from 11.09 until 14.09.2017 in Barcelona. Last year visitor count was about 20000 and Empire of The Sun was rocking the vmworld party :).

I am looking forward to take part at some fascinating speeches, speak with experts and investigate the newest innovations in the vmware universe.

Besides that I am very interested in discoverning the newest products from this years sponsors like Dell EMC,  Intel, Veeam, Fortinet, Nvidia and QNAP.

Twitter will be used during the event.

After the exhibiton you will get a full report about the key facts, also I will try to take some nice pictures.

Maybe there will be an participation in an other big event, stay tuned for another disclosure.

 

Best regards,

Nicolas Frey

Augsburg/Munich town run – 21km

As I mentioned in my about me text, I like to run … fast. Some time ago it was just a meditative recursion but now I am addicted.

My first half marathon in my life turned out to be a success. Placed as 42th of around 1000 on a very nice sunny Sunday in Augsburg, Bavaria. Finishing time: 1:33:45

Last Sunday the 25th of June at 08.00 am in the early morning I participated in the town run in Munich. Over 20000 enthusiastic people took part in this competition. My target was to beat the 1:30h but … not yet. In the end I finished with 1:32:10 which is an improvement nevertheless. The sunglasses were obsolete, very rainy and cold.

 

That’s what my contest preparation looks like:

  • running 3-4 times per week about 20km. Sometimes rope jumping at the end.
  • Split training of various muscles with dumbbells and a professional adjustable bench about 4 times per week. (Ironmaster)
  • 2 times per week stretching and basic yoga exercise.
  • Cycling every day about 10km to work, if the weather tolerates that.
  • Sporadic swimming a bunch of 100meters in lakes or open air baths.
  • Once a week bouldering with friends.
  • Slow juiced vegetable and fruit juice. A balanced and healthy nutrition – most of the time 🙂

 

Best regards,

Nicolas Frey