A logical view on a virtual datacentre services architecture

Posted on By Mark Wilson

A couple of years ago, I wrote a post about a logical view of an End-User Computing (EUC) architecture (which provides a platform for Modern Workplace). It’s served me well and the model continues to be developed (although the changes are subtle so it’s not really worth writing a new post for the 2019 version).

Building on the original EUC/Modern Workplace framework, I started to think what it might look like for datacentre services – and this is something I came up with last year that’s starting to take shape.

Just as for the EUC model, I’ve tried to step up a level from the technology – to get back to the logical building blocks of the solution so that I can apply them according to a specific client’s requirements. I know that it’s far from complete – just look at an Azure or AWS feature list and you can come up with many more classifications for cloud services – but I think it provides the basics and a starting point for a conversation:

Logical view of a virtual datacentre environment

Starting at the bottom left of the diagram, I’ll describe each of the main blocks in turn:

  • Whether hosted on-premises, co-located or making use of public cloud capabilities, Connectivity is a key consideration for datacentre services. This element of the solution includes the WAN connectivity between sites, site-to-site VPN connections to secure access to the datacentre, Internet breakout and network security at the endpoints – specifically the firewalls and other network security appliances in the datacentre.
  • Whilst many of the SBBs in the virtual datacentre services architecture are equally applicable for co-located or on-premises datacentres, there are some specific Cloud Considerations. Firstly, cloud solutions must be designed for failure – i.e. to design out any elements that may lead to non-availability of services (or at least to fail within agreed service levels). Depending on the organisation(s) consuming the services, there may also be considerations around data location. Finally, and most significantly, the cloud provider(s) must practice trustworthy computing and, ideally, will conform to the UK National Cyber Security Centre (NCSC)’s 14 cloud security principles (or equivalent).
  • Just as for the EUC/Modern Workplace architecture, Identity and Access is key to the provision of virtual datacentre services. A directory service is at the heart of the solution, combined with a model for limiting the scope of access to resources. Together with Role Based Access Control (RBAC), this allows for fine-grained access permissions to be defined. Some form of remote access is required – both to access services running in the datacentre and for management purposes. Meanwhile, identity integration is concerned with integrating the datacentre directory service with existing (on-premises) identity solutions and providing SSO for applications, both in the virtual datacentre and elsewhere in the cloud (i.e. SaaS applications).
  • Data Protection takes place throughout the solution – but key considerations include intrusion detection and endpoint security. Just as for end-user devices, endpoint security covers such aspects as firewalls, anti-virus/malware protection and encryption of data at rest.
  • In the centre of the diagram, the Fabric is based on the US National Institute of Standards and Technology (NIST)’s established definition of essential characteristics for cloud computing.
  • The NIST guidance referred to above also defines three service models for cloud computing: Infrastructure as a Service (IaaS); Platform as a Service (PaaS) and Software as a Service (SaaS).
  • In the case of IaaS, there are considerations around the choice of Operating System. Supported operating systems will depend on the cloud service provider.
  • Many cloud service providers will also provide one or more Marketplaces with both first and third-party (ISV) products ranging from firewalls and security appliances to pre-configured application servers.
  • Application Services are the real reason that the virtual datacentre services exist, and applications may be web, mobile or API-based. There may also be traditional hosted server applications – especially where IaaS is in use.
  • The whole stack is wrapped with a suite of Management Tools. These exist to ensure that the cloud services are effectively managed in line with expected practices and cover all of the operational tasks that would be expected for any datacentre including: licensing; resource management; billing; HA and disaster recovery/business continuity; backup and recovery; configuration management; software updates; automation; management policies and monitoring/alerting.

If you have feedback – for example, a glaring hole or suggestions for changes, please feel free to leave a comment below.

Amazon Web Services (AWS) Summit: London Recap

Posted on By Mark Wilson

I’ve written previously about the couple of days I spent at ExCeL in February, learning about Microsoft’s latest developments at the Ignite Tour and, a few weeks later I found myself back at the same venue, this time focusing on Amazon Web Services (AWS) at the London AWS Summit (four years since my last visit).

Even with a predominantly Microsoft-focused client base, there are situations where a multi-cloud solution is required and so, it makes sense for me to expand my knowledge to include Amazon’s cloud offerings. I may not have the detail and experience that I have with Microsoft Azure, but certainly enough to make an informed choice within my Architect role.

One of the first things I noticed is that, for Amazon, it’s all about the numbers. The AWS Summit had a lot of attendees – 12000+ were claimed, for more than 60 technical sessions supported by 98 sponsoring partners. Frankly, it felt to me that there were a few too many people there at times…

AWS is clearly growing – citing 41% growth comparing Q1 2019 with Q1 2018. And, whilst the comparisons with the industrial revolution and the LSE research that shows 95% of today’s startups would find traditional IT models limiting today were all good and valid, the keynote soon switched to focus on AWS claims of “more”. More services. More depth. More breadth.

There were some good customer slots in the keynote: Sainsbury’s Group CIO Phil Jordan and Group Digital Officer Clodagh Moriaty spoke about improving online experiences, integrating brands such as Nectar and Sainsbury’s, and using machine learning to re-plan retail space and to plan online deliveries. Ministry of Justice CDIO Tom Read talked about how the MOJ is moving to a microservice-based application architecture.

After the keynote, I immersed myself in technical sessions. In fact, I avoided the vendor booths completely because the room was absolutely packed when I tried to get near. My afternoon consisted of:

  • Driving digital transformation using artificial intelligence by Steven Bryen (@Steven_Bryen) and Bjoern Reinke.
  • AWS networking fundamentals by Perry Wald and Tom Adamski.
  • Creating resilience through destruction by Adrian Hornsby (@adhorn).
  • How to build an Alexa Skill in 30 minutes by Andrew Muttoni (@muttonia).

All of these were great technical sessions – and probably too much for a single blog post but, here goes anyway…

Driving digital transformation using artificial intelligence

Amazon thinks that driving better customer experience requires Artificial Intelligence (AI), specifically Machine Learning (ML). Using an old picture of London Underground workers sorting through used tickets in the 1950s to identify the most popular journeys, Steven Bryen suggested that more data leads to better analytics and better outcomes that can be applied in more ways (in a cyclical manner).

The term “artificial intelligence” has been used since John McCarthy coined it in 1955. The AWS view is that AI taking off because of:

  • Algorithms.
  • Data (specifically the ability to capture and store it at scale).
  • GPUs and acceleration.
  • Cloud computing.

Citing research from PwC [which I can’t find on the Internet], AWS claim that world GDP was $80Tn in 2018 and is expected to be $112Tn in 2030  ($15.7Tn of which can be attributed to AI).

Data science, artificial intelligence, machine learning and deep learning can be thought of as a series of concentric rings.

Machine learning can be supervised learning (betting better at finding targets); unsupervised (assume nothing and question everything); or reinforcement learning (rewarding high performing behaviour).

Amazon claims extensive AI experience through its own ML experience:

  • Recommendations Engine
  • Prime Air
  • Alexa
  • Go (checkoutless stores)
  • Robotic warehouses – taking trolleys to packer to scan and pack (using an IoT wristband to make sure robots avoid maintenance engineers).

Every day Amazon applies new AI/ML-based improvements to its business, at a global scale through AWS.

Challenges for organisations are that:

  • ML is rare
  • plus: Building and scaling ML technology is hard
  • plus: Deploying and operating models in production is time-consuming and expensive
  • equals: a lack of cost-effective easy-to-use and scalable ML services

Most time is spent getting data ready to get intelligence from it. Customers need a complete end-to-end ML stack and AWS provides that with edge technologies such as Greengrass for offline inference and modelling in SageMaker. The AWS view is that ML prediction becomes a RESTful API call.

With the scene set, Steven Bryen handed over to Bjoern Reinke, Drax Retail’s Director of Smart Metering.

Drax has converted former coal-fired power stations to use biomass: capturing carbon into biomass pellets, which are burned to create steam that drives turbines – representing 15% of the UK’s renewable energy.

Drax uses a systems thinking approach with systems of record, intelligence and engagement

System of intelligence need:

  • Trusted data.
  • Insight everywhere.
  • Enterprise automation.

Customers expect tailoring: efficiency; security; safety; and competitive advantage.

Systems of intelligence can be applied to team leaders, front line agents (so they already know that customer has just been online looking for a new tariff), leaders (for reliable data sources), and assistant-enabled recommendations (which are no longer futuristic).

Fragmented/conflicting data is pumped into a data lake from where ETL and data warehousing technologies are used for reporting and visualisation. But Drax also pull from the data lake to run analytics for data science (using Inawisdom technology).

The data science applications can monitor usage and see base load, holidays, etc. Then, they can look for anomalies – a deviation from an established time series. This might help to detect changes in tenants, etc. and the information can be surfaced to operations teams.

AWS networking fundamentals

After hearing how AWS can be used to drive insight into customer activities, the next session was back to pure tech. Not just tech but infrastructure (all be it as a service). The following notes cover off some AWS IaaS concepts and fundamentals.

Customers deploy into virtual private cloud (VPC) environments within AWS:

  • For demonstration purposes, a private address range (CIDR) was used – 172.31.0.0/16 (a private IP range from RFC 1918). Importantly, AWS ranges should be selected to avoid potential conflicts with on-premises infrastructure. Amazon recommends using /16 (65536 addresses) but network teams may suggest something smaller.
  • AWS is dual-stack (IPv4 and IPv6) so even if an IPv6 CIDR is used, infrastructure will have both IPv4 and IPv6 addresses.
  • Each VPC should be broken into availability zones (AZs), which are risk domains on different power grids/flood profiles and a subnet placed in each (e.g. 172.31.0.0/24, 172.31.1.0/24, 172.31.2.0/24).
  • Each VPC has a default routing table but an administrator can create and assign different routing tables to different subnets.

To connect to the Internet you will need a connection, a route and a public address:

  • Create a public subnet (one with public and private IP addresses).
  • Then, create an Internet Gateway (IGW).
  • Finally, Create a route so that the default gateway is the IGW (172.31.0.0/16 local and 0.0.0.0/0 igw_id).
  • Alternatively, create a private subnet and use a NAT gateway for outbound only traffic and direct responses (172.31.0.0/16 local and 0.0.0.0/0 nat_gw_id).

Moving on to network security:

  • Network Security Groups (NSGs) provide a stateful distributed firewall so a request from one direction automatically sets up permissions for a response from the other (avoiding the need to set up separate rules for inbound and outbound traffic).
    • Using an example VPC with 4 web servers and 3 back end servers:
      • Group into 2 security groups
      • Allow web traffic from anywhere to web servers (port 80 and source 0.0.0.0/0)
      • Only allow web servers to talk to back end servers (port 2345 and source security group ID)
  • Network Access Control Lists (NACLs) are stateless – they are just lists and need to be explicit to allow both directions.
  • Flow logs work at instance, subnet or VPC level and write output to S3 buckets or CloudWatch logs. They can be used for:
    • Visibility
    • Troubleshooting
    • Analysing traffic flow (no payload, just metadata)
      • Network interface
      • Source IP and port
      • Destination IP and port
      • Bytes
      • Condition (accept/reject)
  • DNS in a VPC is switched on by default for resolution and assigning hostnames (rather than just using IP addresses).
    • AWS also has the Route 53 service for customers who would like to manage their own DNS.

Finally, connectivity options include:

  • Peering for private communication between VPCs
    • Peering is 1:1 and can be in different regions but the CIDR must not overlap
    • Each VPC owner can send a request which is accepted by the owner on the other side. Then, update the routing tables on the other side.
    • Peering can get complex if there are many VPCs. There is also a limit of 125 peerings so a Transit Gateway can be used to act as a central point but there are some limitations around regions.
    • Each Transit Gateway can support up to 5000 connections.
  • AWS can be connected to on-premises infrastructure using a VPN or with AWS Direct Connect
    • A VPN is established with a customer gateway and a virtual private gateway is created on the VPC side of the connection.
      • Each connection has 2 tunnels (2 endpoints in different AZs).
      • Update the routing table to define how to reach on-premises networks.
    • Direct Connect
      • AWS services on public address space are outside the VPC.
      • Direct Connect locations have a customer or partner cage and an AWS cage.
      • Create a private virtual interface (VLAN) and a public virtual interface (VLAN) for access to VPC and to other AWS services.
      • A Direct Connect Gateway is used to connect to each VPC
    • Before Transit Gateway customers needed a VPN per VPC.
      • Now they can consolidate on-premises connectivity
      • For Direct Connect it’s possible to have a single tunnel with a Transit Gateway between the customer gateway and AWS.
  • Route 53 Resolver service can be used for DNS forwarding on-premises to AWS and vice versa.
  • VPC Sharing provides separation of resources with:
    • An Owner account to set up infrastructure/networking.
    • Subnets shared with other AWS accounts so they can deploy into the subnet.
  • Interface endpoints make an API look as if it’s part of an organisation’s VPC.
    • They override the public domain name for service.
    • Using a private link can only expose a specific service port and control the direction of communications and no longer care about IP addresses.
  • Amazon Global Accelerator brings traffic onto the AWS backbone close to end users and then uses that backbone to provide access to services.

Creating resilience through destruction

Adrian Horn presenting at AWS Summit London

One of the most interesting sessions I saw at the AWS Summit was Adrian Horn’s session that talked about deliberately breaking things to create resilience – which is effectively the infrastructure version of test-driven development (TDD), I guess…

Actually, Adrian made the point that it’s not so much the issues that bringing things down causes as the complexity of bringing them back up.

“Failures are a given and everything will eventually fail over time”

Werner Vogels, CTO, Amazon.com

We may break a system into microservices to scale but we also need to think about resilience: the ability for a system to handle and eventually recover from unexpected conditions.

This needs to consider a stack that includes:

  • People
  • Application
  • Network and Data
  • Infrastructure

And building confidence through testing only takes us so far. Adrian referred to another presentation, by Jesse Robbins, where he talks about creating resilience through destruction.

Firefighters train to build intuition – so they know what to do in the event of a real emergency. In IT, we have the concept of chaos engineering – deliberately injecting failures into an environment:

  • Start small and build confidence:
    • Application level
    • Host failure
    • Resource attacks (CPU, latency…)
    • Network attacks (dependencies, latency…)
    • Region attack
    • Human attack (remove a key resource)
  • Then, build resilient systems:
    • Steady state
    • Hypothesis
    • Design and run an experiment
    • Verify and learn
    • Fix
    • (maybe go back to experiment or to start)
  • And use bulkheads to isolate parts of the system (as in shipping).

Think about:

  • Software:
    • Certificate Expiry
    • Memory leaks
    • Licences
    • Versioning
  • Infrastructure:
    • Redundancy (multi-AZ)
    • Use of managed services
    • Bulkheads
    • Infrastructure as code
  • Application:
    • Timeouts
    • Retries with back-offs (not infinite retries)
    • Circuit breakers
    • Load shedding
    • Exception handing
  • Operations:
    • Monitoring and observability
    • Incident response
    • Measure, measure, measure
    • You build it, your run it

AWS’ Well Architected framework has been developed to help cloud architects build secure, high-performing, resilient, and efficient infrastructure for their applications, based on some of these principles.

Adrian then moved on to consider what a steady state looks like:

  • Normal behaviour of system
  • Business metric (e.g. pulse of Netflix – multiple clicks on play button if not working)
    • Amazon extra 100ms load time led to 1% drop in sales (Greg Linden)
    • Google extra 500ms of load time led to 20% fewer searches (Marissa Mayer)
    • Yahoo extra 400ms of load time caused 5-9% increase in back clicks (Nicole Sullivan)

He suggests asking questions about “what if?” and following some rules of thumb:

  • Start very small
  • As close as possible to production
  • Minimise the blast radius
  • Have an emergency stop
    • Be careful with state that can’t be rolled back (corrupt or incorrect data)

Use canary deployment with A-B testing via DNS or similar for chaos experiment (1%) or normal (99%).

Adrian then went on to demonstrate his approach to chaos engineering, including:

  • Fault injection queries for Amazon Aurora (can revert immediately)
    • Crash a master instance
    • Fail a replica
    • Disk failure
    • Disk congestion
  • DDoS yourself
  • Add latency to network
    • ~ tc qdisc add dev eth0 root netem delay 200ms
  • https://github.com/Netflix/SimianArmy
    • Shut down services randomly
    • Slow down performance
    • Check conformity
    • Break an entire region
    • etc.
  • The chaos toolkit
  • Gremin
    • Destruction as a service!
  • ToxiProxy
    • Sit between components and add “toxics” to test impact of issues
  • Kube-Money project (for Kubernetes)
  • Pumba (for Docker)
  • Thundra (for Lambda)

Use post mortems for correction of errors – the 5 whys. Also, understand that there is no isolated “cause” of an accident.

My notes don’t do Adrian’s talk justice – there’s so much more that I could pick up from re-watching his presentation. Adrian tweeted a link to his slides and code – if you’d like to know more, check them out:

How to build an Alexa Skill in 30 minutes

Spoiler: I didn’t have a working Alexa skill at the end of my 30 minutes… nevertheless, here’s some info to get you started!

Amazon’s view is that technology tries to constrain us. Things got better with mobile and voice is the next step forward. With voice, we can express ourselves without having to understand a user interface [except we do, because we have to know how to issue commands in a format that’s understood – that’s the voice UI!].

I get the point being made – to add an item to a to-do list involves several steps:

  • Find phone
  • Unlock phone
  • Find app
  • Add item
  • etc.

Or, you could just say (for example) “Alexa, ask Ocado to add tuna to my trolley”.

Alexa is a service in the AWS cloud that understands request and acts upon them. There are two components:

  • Alexa voice service – how a device manufacturer adds Alexa to its products.
  • Alexa Skills Kit – to create skills that make something happen (and there are currently more than 80,000 skills available).

An Alexa-enabled device only needs to know to wake up, then stream some “mumbo jumbo” to the cloud, at which point:

  • Automatic speech recognition with translate text to speech
  • Natural language understanding will infer intent (not just text, but understanding…)

Creating skills is requires two parts:

Alexa-hosted skills use Lambda under the hood and creating the skill involves:

  1. Give the skill a name.
  2. Choose the development model.
  3. Choose a hosting method.
  4. Create a skill.
  5. Test in a simulation environment.

Finally, some more links that may be useful:

In summary

Looking back, the technical sessions made my visit to the AWS Summit worthwhile but overall, I was a little disappointed, as this tweet suggests:

Would I recommend the AWS Summit to others? Maybe. Would I watch the keynote from home? No. Would I try to watch some more technical sessions? Absolutely, if they were of the quality I saw on the day. Would I bother to go to ExCeL with 12000 other delegates herded like cattle? Probably not…

Shop local when buying a new bike

Posted on By Mark Wilson

Back in 2013, I bought myself a road bike. It’s a Bianchi Via Nirone 7 C2C and it was the first road bike I’d had since my teenage years when I had a 21-speed “racer” (complete with shifters on the down tube).

My Bianchi has served me well but, after nearly 12000km I’m starting to notice some hairline cracks in the paint, a bit of corrosion on the chainstay – and I recently had to cut out one of the upgrades I’d made as the carbon fibre seat post had bonded itself to the inside of the aluminium alloy seat tube.

I’d been saving up for a new bike for a while (promising myself that I could have a new bike when I lost some weight…) but I decided to retire the Bianchi (or at least just use it for Zwifting) and get something new (maybe I can lose some weight by riding more now I have the new bike).

For a long while, I was tempted by a Canyon Endurace CF SL Disc 8.0 Di2. Canyon make some lovely bikes but they are mail-order only (unless you can visit them in Germany). Not having distributors reduces the price, but it also increases the risk of buying the wrong size, etc. added to which, recent experience (buying a frame from Planet X for my son) showed me that sometimes you get what you pay for.

I also feel guilty every time I shop at Wiggle – we’ll miss our local bike shops (LBS) when they are gone and I’ve relied on a few for parts at short notice recently (including Corley Cycles and Chaineys in Milton Keynes). But, just like buying from Amazon instead of a high street store, sometimes the economics mean it just makes sense. Even so, with a new bike purchase, I wasn’t entirely comfortable buying online.

I looked at some of the other mainstream brands too (how about a Trek Domane?). But what about the price difference?

Well, there were a few things to take into account there:

  • Online sizing tools are good, but not perfect and the Canyon would need a bike fit before I could be sure I was ordering the right size. Corley Cycles included not only the sizing fit but also an advanced bike fit with the new bike.
  • Then, membership of my local cycling club got me a further discount (10%).
  • At this point, we’re getting close to pretty much the same price.
  • Chuck in some bottles, cages, and a lot of advice – plus I’m helping to keep my LBS in business and I decided that I’d rather have the “purchase from a shop” experience.

So, I’m now the proud owner of the new Specialized Roubaix Comp (2020 edition). Sure, the lightweight endurance bike with electronic shifting became a lightweight endurance bike with mechanical shifting and front suspension instead but my conscience is clear – and it is pretty damned awesome.

WB-40 appearances

Posted on By Mark Wilson

Several years ago, I met Matt Ballantine (@ballantine70), when he was working at Microsoft. Over the years, we’ve had many conversations in person, online and over social media and I’ve been listening to his WB-40 podcast with Chris Weston (@chrisweston) since they started it in 2016.

WB-40 Podcast logo

In recent months, I’ve been fortunate to feature a few times on the podcast:

Appearing on WB-40 (and on The Flexible Movement) has made me think a little about maybe starting a podcast of my own. James Bannan (@JamesBannan) and I had a podcast called Coalface Tech for a while a few years ago but we found working on opposite sides of the planet and recording decent audio challenging at the time. At the moment I struggle to write blog posts so, let’s see if that ever gets off the drawing board.

In the meantime, if you’re interested in the intersection of IT and business then I recommend checking out WB-40, which will also transition from its online form to the physical world next week, with a live event in London.

The perils of mixing aluminium alloy and carbon fibre cycle components (removing a stuck seatpost)

Posted on By Mark Wilson

When I first started road cycling, a friend suggested I could increase my comfort for not too much money with a carbon seatpost upgrade. A short while later I was the proud owner of a Deda Superzero post and it looked good, even if the marginal gains with me riding the bike were so small that they could be offset by an extra cup of coffee…

Fast forward a few years and my teenage son was borrowing my bike as he’d outgrown his. I went to adjust the seatpost and found it was stuck fast. And I mean stuck. Solid.

Obviously, this was fine for me to ride the bike but I couldn’t just leave it. Over time aluminium alloy (e.g. my frame) will bond itself to carbon fibre (e.g. my seatpost) and even though a bike mechanic had applied some anti-seize paste and I’d given the post a “wiggle” from time to time, I’d obviously left it too long since the last check.

So I hit the Internet and Googled “how to remove stuck seat post carbon aluminium”…

There’s a load of ideas on the Sheldon Brown bicycle technical info site.

And this thread at BikeRadar is useful too.

Added to which, there’s more info at Cycling UK including a health warning about cutting carbon fibre. The dust is nasty stuff…

I’d used penetrating oils and hot water on the frame but the bond was too tight – not enough space to get the oil down. So, with assistance from a fellow club member who was very generous with his time and use of his tools, I tried heating and cooling the affected areas of frame and post with a heat gun and some “Shock and Unlock”. Ideally, this would cool the post and heat the frame – the idea being that one will expand and the other contract, breaking the bond. Nothing.

Heating and cooling the seatpost Frozen seatpost

Then we tried the other way – even if it pushed the materials together it might break the bond on cooling. This was quickly followed by disaster – I attempted to move the post when it had been heated, it seemed to move, too easily and I found I’d actually twisted it and deformed the post.

Snapped seatpost

“Right, now that post is a write-off, at least the frame is OK, we’ll have to cut it out.”

We tried using naked hacksaw blades (wet seems to help and I also seemed to do better with 36T blades than 24T), a padsaw, even an electric reciprocating saw.

Cutting out the seatpost

Eventually (and I mean after many hours) I had cut some channels in the remains of the post but it still wasn’t coming out. The idea is that it should collapse in on itself once there’s a vertical cut that stops it pushing against the frame seat tube. What did happen though was that the carbon fibre delaminated and we were able to chisel pieces out using a variety of punches chisels and screwdrivers and a lot of GT85 (being careful not to damage the frame).

Eventually, the post came out, in thousands of pieces over a few more hours. And the depressing part, when the bottom section of the seatpost was removed (the last few inches hadn’t seized), we could see just how little effect the hacksaw had made that far down the post – there just wasn’t enough force to make a difference.

With the last of the pieces out I cleaned up the inside of the frame with some wire wool (using a bent coat hanger to pull it back up) and put the original aluminium seatpost back. There are some minor scrapes inside the seat tube and some paint cracks on the top section too so I’m not sure whether I’ve compromised the frame.

Seatpost removed

Thankfully, I brought forward my plans for a new bike, bought my son a larger frame for his and this one is now in semi-retirement, relegated to Zwift duties.

It took me (and a friend) at least 8, maybe 12 hours to remove that post so, if you do mix aluminium alloy and carbon fibre components, make sure you (re)move them regularly. 

Microsoft Ignite | The Tour: London Recap

Posted on By Mark Wilson

One of the most valuable personal development activities in my early career was a trip to the Microsoft TechEd conference in Amsterdam. I learned a lot – not just technically but about making the most of events to gather information, make new industry contacts, and generally top up my knowledge. Indeed, even as a relatively junior consultant, I found that dipping into multiple topics for an hour or so gave me a really good grounding to discover more (or just enough to know something about the topic) – far more so than an instructor-led training course.

Over the years, I attended further “TechEd”s in Amsterdam, Barcelona and Berlin. I fought off the “oh Mark’s on another jolly” comments by sharing information – incidentally, conference attendance is no “jolly” – there may be drinks and even parties but those are after long days of serious mental cramming, often on top of broken sleep in a cheap hotel miles from the conference centre.

Microsoft TechEd is no more. Over the years, as the budgets were cut, the standard of the conference dropped and in the UK we had a local event called Future Decoded. I attended several of these – and it was at Future Decoded that I discovered risual – where I’ve been working for almost four years now.

Now, Future Decoded has also fallen by the wayside and Microsoft has focused on taking it’s principal technical conference – Microsoft Ignite – on tour, delivering global content locally.

So, a few weeks ago, I found myself at the ExCeL conference centre in London’s Docklands, looking forward to a couple of days at “Microsoft Ignite | The Tour: London”.

Conference format

Just like TechEd, and at Future Decoded (in the days before I had to use my time between keynotes on stand duty!), the event was broken up into tracks with sessions lasting around an hour. Because that was an hour of content (and Microsoft event talks are often scheduled as an hour, plus 15 minutes Q&A), it was pretty intense, and opportunities to ask questions were generally limited to trying to grab the speaker after their talk, or at the “Ask the Experts” stands in the main hall.

One difference to Microsoft conferences I’ve previously attended was the lack of “level 400” sessions: every session I saw was level 100-300 (mostly 200/300). That’s fine – that’s the level of content I would expect but there may be some who are looking for more detail. If it’s detail you’re after then Ignite doesn’t seem to be the place.

Also, I noticed that Day 2 had fewer delegates and lacked some of the “hype” from Day 1: whereas the Day 1 welcome talk was over-subscribed, the Day 2 equivalent was almost empty and light on content (not even giving airtime to the conference sponsors). Nevertheless, it was easy to get around the venue (apart from a couple of pinch points).

Personal highlights

I managed to cover 11 topics over two days (plus a fair amount of networking). The track format of the event was intended to let a delegate follow a complete learning path but, as someone who’s a generalist (that’s what Architects have to be), I spread myself around to cover:

  • Dealing with a massive onset of data ingestion (Jeramiah Dooley/@jdooley_clt).
  • Enterprise network connectivity in a cloud-first world (Paul Collinge/@pcollingemsft).
  • Building a world without passwords.
  • Discovering Azure Tooling and Utilities (Simona Cotin/@simona_cotin).
  • Selecting the right data storage strategy for your cloud application (Jeramiah Dooley/@jdooley_clt).
  • Governance in Azure (Sam Cogan/@samcogan).
  • Planning and implementing hybrid network connectivity (Thomas Maurer/@ThomasMaurer).
  • Transform device management with Windows Autopilot, Intune and OneDrive (Michael Niehaus/@mniehaus and Mizanur Rahman).
  • Maintaining your hybrid environment (Niel Peterson/@nepeters).
  • Windows Server 2019 Deep Dive (Jeff Woolsey/@wsv_guy).
  • Consolidating infrastructure with the Azure Kubernetes Service (Erik St Martin/@erikstmartin).

In the past, I’d have written a blog post for each topic. I was going to say that I simply don’t have the time to do that these days but by the time I’d finished writing this post, I thought maybe I could have split it up a bit more! Regardless, here are some snippets of information from my time at Microsoft Ignite | The Tour: London. There’s more information in the slide decks – which are available for download, along with the content for the many sessions I didn’t attend.

Data ingestion

Ingesting data can be broken into:

  • Real-time ingestion.
  • Real-time analysis (see trends as they happen – and make changes to create a competitive differentiator).
  • Producing actions as patterns emerge.
  • Automating reactions in external services.
  • Making data consumable (in whatever form people need to use it).

Azure has many services to assist with this – take a look at IoT Hub, Azure Event Hubs, Azure Databricks and more.

Enterprise network connectivity for the cloud

Cloud traffic is increasing whilst traffic that remains internal to the corporate network is in decline. Traditional management approaches are no longer fit for purpose.

Office applications use multiple persistent connections – this causes challenges for proxy servers which generally degrade the Office 365 user experience. Remediation is possible, with:

  • Differentiated traffic – follow Microsoft advice to manage known endpoints, including the Office 365 IP address and URL web service.
  • Let Microsoft route traffic (data is in a region, not a place). Use DNS resolution to egress connections close to the user (a list of all Microsoft peering locations is available). Optimise the route length and avoid hairpins.
  • Assess network security using application-level security, reducing IP ranges and ports and evaluating the service to see if some activities can be performed in Office 365, rather than at the network edge (e.g. DLP, AV scanning).

For Azure:

  • Azure ExpressRoute is a connection to the edge of the Microsoft global backbone (not to a datacentre). It offers 2 lines for resilience and two peering types at the gateway – private and public (Microsoft) peering.
  • Azure Virtual WAN can be used to build a hub for a region and to connect sites.
  • Replace branch office routers with software-defined (SDWAN) devices and break out where appropriate.
Microsoft global network

Passwordless authentication

Basically, there are three options:

  • Windows Hello.
  • Microsoft Authenticator.
  • FIDO2 Keys.

Azure tooling and utilities

Useful resources include:

Selecting data storage for a cloud application

What to use? It depends! Classify data by:

  • Type of data:
    • Structured (fits into a table)
    • Semi-structured (may fit in a table but may also use outside metadata, external tables, etc.)
    • Unstructured (documents, images, videos, etc.)
  • Properties of the data:
    • Volume (how much)
    • Velocity (change rate)
    • Variety (sources, types, etc.)
Item TypeVolume Velocity Variety
Product catalogue Semi-structured High Low Low
Product photos Unstructured High Low Low
Sales data Semi-structured Medium High High

How to match data to storage:

  • Storage-driven: build apps on what you have.
  • Cloud-driven: deploy to the storage that makes sense.
  • Function-driven: build what you need; storage comes with it.

Governance in Azure

It’s important to understand what’s running in an Azure subscription – consider cost, security and compliance:

  • Review (and set a baseline):
    • Tools include: Resource Graph; Cost Management; Security Center; Secure Score.
  • Organise (housekeeping to create a subscription hierarchy, classify subscriptions and resources, and apply access rights consistently):
    • Tools include: Management Groups; Tags; RBAC;
  • Audit:
    • Make changes to implement governance without impacting people/work. Develop policies, apply budgets and audit the impact of the policies.
    • Tools include: Cost Management; Azure Policy.
  • Enforce
    • Change policies to enforcement, add resolution actions and enforce budgets.
    • Consider what will happen for non-compliance?
    • Tools include: Azure Policy; Cost Management; Azure Blueprints.
  • (Loop back to review)
    • Have we achieved what we wanted to?
    • Understand what is being spent and why.
    • Know that only approved resources are deployed.
    • Be sure of adhering to security practices.
    • Opportunities for further improvement.

Planning and implementing hybrid network connectivity

Moving to the cloud allows for fast deployment but planning is just as important as it ever was. Meanwhile, startups can be cloud-only but most established organisations have some legacy and need to keep some workloads on-premises, with secure and reliable hybrid communication.

Considerations include:

  • Extension of the internal protected network:
    • Should workloads in Azure only be accessible from the Internal network?
    • Are Azure-hosted workloads restricted from accessing the Internet?
    • Should Azure have a single entry and egress point?
    • Can the connection traverse the public Internet (compliance/regulation)?
  • IP addressing:
    • Existing addresses on-premises; public IP addresses.
    • Namespaces and name resolution.
  • Multiple regions:
    • Where are the users (multiple on-premises sites); where are the workloads (multiple Azure regions); how will connectivity work (should each site have its own connectivity)?
  • Azure virtual networks:
    • Form an isolated boundary with secure communications.
    • Azure-assigned IP addresses (no need for a DHCP server).
    • Segmented with subnets.
    • Network Security Groups (NSGs) create boundaries around subnets.
  • Connectivity:
    • Site to site (S2S) VPNs at up to 1Gbps
      • Encrypted traffic over the public Internet to the GatewaySubnet in Azure, which hosts VPN Gateway VMs.
      • 99.9% SLA on the Gateway in Azure (not the connection).
      • Don’t deploy production workloads on the GatewaySubnet; /26, /27 or /28 subnets recommended; don’t apply NSGs to the GatewaySubnet – i.e. let Azure manage it.
    • Dedicated connections (Azure ExpressRoute): private connection at up to 10Gbps to Azure with:
      • Private peering (to access Azure).
      • Microsoft peering (for Office 365, Dynamics 365 and Azure public IPs).
      • 99.9% SLA on the entire connection.
    • Other connectivity services:
      • Azure ExpressRoute Direct: a 100Gbps direct connection to Azure.
      • Azure ExpressRoute Global Reach: using the Microsoft network to connect multiple local on-premises locations.
      • Azure Virtual WAN: branch to branch and branch to Azure connectivity with software-defined networks.
  • Hybrid networking technologies:

Modern Device Management (Autopilot, Intune and OneDrive)

The old way of managing PC builds:

  1. Build an image with customisations and drivers
  2. Deploy to a new computer, overwriting what was on it
  3. Expensive – and the device has a perfectly good OS – time-consuming

Instead, how about:

  1. Unbox PC
  2. Transform with minimal user interaction
  3. Device is ready for productive use

The transformation is:

  • Take OEM-optimised Windows 10:
    • Windows 10 Pro and drivers.
    • Clean OS.
  • Plus software, settings, updates, features, user data (with OneDrive for Business).
  • Ready for productive use.

The goal is to reduce the overall cost of deploying devices. Ship to a user with half a page of instructions…

Windows Autopilot overview

Autopilot deployment is cloud driven and will eventually be centralised through Intune:

  1. Register device:
    • From OEM or Channel (manufacturer, model and serial number).
    • Automatically (existing Intune-managed devices).
    • Manually using a PowerShell script to generate a CSV file with serial number and hardware hash, which is then uploaded to the Intune portal.
  2. Assign Autopilot profile:
    • Use Azure AD Groups to assign/target.
    • The profile includes settings such as deployment mode, BitLocker encryption, device naming, out of box experience (OOBE).
    • An Azure AD device object is created for each imported Autopilot device.
  3. Deploy:
    • Needs Azure AD Premium P1/P2
    • Scenarios include:
      • User-driven with Azure AD:
        • Boot to OOBE, choose language, locale, keyboard and provide credentials.
        • The device is joined to Azure AD, enrolled to Intune and policies are applied.
        • User signs on and user-assigned items from Intune policy are applied.
        • Once the desktop loads, everything is present, including file links in OneDrive) – time depends on the software being pushed.
      • Self-deploying (e.g. kiosk, digital signage):
        • No credentials required; device authenticates with Azure AD using TPM 2.0.
      • User-driven with hybrid Azure AD join:
        • Requires Offline Domain Join Connector to create AD DS computer account.
        • Device connected to the corporate network (in order to access AD DS), registered with Autopilot, then as before.
        • Sign on to Azure AD and then to AD DS during deployment. If they use the same UPN then it makes things simple for users!
      • Autopilot for existing devices (Windows 7 to 10 upgrades):
        • Backup data in advance (e.g. with OneDrive)
        • Deploy generic Windows 10.
        • Run Autopilot user-driven mode (can’t harvest hardware hashes in Windows 7 so use a JSON config file in the image – the offline equivalent of a profile. Intune will ignore unknown device and Autopilot will use the file instead; after deployment of Windows 10, Intune will notice a PC in the group and apply the profile so it will work if the PC is reset in future).

Autopilot roadmap (1903) includes:

  • “White glove” pre-provisioning for end users: QR code to track, print welcome letter and shipping label!
  • Enrolment status page (ESP) improvements.
  • Cortana voiceover disabled on OOBE.
  • Self-updating Autopilot (update Autopilot without waiting to update Windows).

Maintaining your hybrid environment

Common requirements in an IaaS environment include wanting to use a policy-based configuration with a single management and monitoring solution and auto-remediation.

Azure Automation allows configuration and inventory; monitoring and insights; and response and automation. The Azure Portal provides a single pane of glass for hybrid management (Windows or Linux; any cloud or on-premises).

For configuration and state management, use Azure Automation State Configuration (built on PowerShell Desired State Configuration).

Inventory can be managed with Log Analytics extensions for Windows or Linux. An Azure Monitoring Agent is available for on-premises or other clouds. Inventory is not instant though – can take 3-10 minutes for Log Analytics to ingest the data. Changes can be visualised (for state tracking purposes) in the Azure Portal.

Azure Monitor and Log Analytics can be used for data-driven insights, unified monitoring and workflow integration.

Responding to alerts can be achieved with Azure Automation Runbooks, which store scripts in Azure and run them in Azure. Scripts can use PowerShell or Python so support both Windows and Linux). A webhook can be triggered with and HTTP POST request. A Hybrid runbook worker can be used to run on-premises or in another cloud.

It’s possible to use the Azure VM agent to run a command on a VM from Azure portal without logging in!

Windows Server 2019

Windows Server strategy starts with Azure. Windows Server 2019 is focused on:

  • Hybrid:
    • Backup/connect/replicate VMs.
    • Storage Migration Service to migrate unstructured data into Azure IaaS or another on-premises location (from 2003+ to 2016/19).
      1. Inventory (interrogate storage, network security, SMB shares and data).
      2. Transfer (pairings of source and destination), including ACLs, users and groups. Details are logged in a CSV file.
      3. Cutover (make the new server look like the old one – same name and IP address). Validate before cutover – ensure everything will be OK. Read-only process (except change of name and IP at the end for the old server).
    • Azure File Sync: centralise file storage in Azure and transform existing file servers into hot caches of data.
    • Azure Network Adapter to connect servers directly to Azure networks (see above).
  • Hyper-converged infrastructure (HCI):
    • The server market is still growing and is increasingly SSD-based.
    • Traditional rack looked like SAN, storage fabric, hypervisors, appliances (e.g. load balancer) and top of rack Ethernet switches.
    • Now we use standard x86 servers with local drives and software-defined everything. Manage with Admin Center in Windows Server (see below).
    • Windows Server now has support for persistent memory: DIMM-based; still there after a power-cycle.
    • The Windows Server Software Defined (WSSD) programme is the Microsoft approach to software-defined infrastructure.
  • Security: shielded VMs for Linux (VM as a black box, even for an administrator); integrated Windows Defender ATP; Exploit Guard; System Guard Runtime.
  • Application innovation: semi-annual updates are designed for containers. Windows Server 2019 is the latest LTSC channel so it has the 1709/1803 additions:
    • Enable developers and IT Pros to create cloud-native apps and modernise traditional apps using containers and micro services.
    • Linux containers on Windows host.
    • Service Fabric and Kubernetes for container orchestration.
    • Windows subsystem for Linux.
    • Optimised images for server core and nano server.

Windows Admin Center is core to the future of Windows Server management and, because it’s based on remote management, servers can be core or full installations – even containers (logs and console). Download from http://aka.ms/WACDownload

  • 50MB download, no need for a server. Runs in a browser and is included in Windows/Windows Server licence
  • Runs on a layer of PowerShell. Use the >_ icon to see the raw PowerShell used by Admin Center (copy and paste to use elsewhere).
  • Extensible platform.

What’s next?

  • More cloud integration
  • Update cadence is:
    • Insider builds every 2 weeks.
    • Semi-annual channel every 6 months (specifically for containers):
      • 1709/1803/1809/19xx.
    • Long-term servicing channel
      • Every 2-3 years.
      • 2016, 2019 (in September 2018), etc.

Windows Server 2008 and 2008 R2 reach the end of support in January 2020 but customers can move Windows Server 2008/2008 R2 servers to Azure and get 3 years of security updates for free (on-premises support is chargeable).

Further reading: What’s New in Windows Server 2019.

Containers/Azure Kubernetes Service

Containers:

  • Are fully-packaged applications that use a standard image format for better resource isolation and utilisation.
  • Are ready to deploy via an API call.
  • Are not Virtual machines (for Linux).
  • Do not use hardware virtualisation.
  • Offer no hard security boundary (for Linux).
  • Can be more cost effective/reliable.
  • Have no GUI.

Kubernetes is:

  • An open source system for auto-deployment, scaling and management of containerized apps.
  • Container Orchestrator to manage scheduling; affinity/anti-affinity; health monitoring; failover; scaling; networking; service discovery.
  • Modular and pluggable.
  • Self-healing.
  • Designed by Google based on a system they use to run billions of containers per week.
  • Described in “Phippy goes to the zoo”.

Azure container offers include:

  • Azure Container Instances (ACI): containers on demand (Linux or Windows) with no need to provision VMs or clusters; per-second billing; integration with other Azure services; a public IP; persistent storage.
  • Azure App Service for Linux: a fully-managed PaaS for containers including workflows and advanced features for web applications.
  • Azure Kubernetes Service (AKS): a managed Kubernetes offering.

Wrap-up

So, there you have it. An extremely long blog post with some highlights from my attendance at Microsoft Ignite | The Tour: London. It’s taken a while to write up so I hope the notes are useful to someone else!

Strava Art

Posted on By Mark Wilson

It’s no secret that I enjoy cycling – and I’m also a bit of a geek. My cycling and my tech come together in various places but Strava is one of the most obvious… I have been known to say “If it’s not on Strava, it didn’t happen” (of course, that’s in jest – but I do get annoyed if my GPS traces get messed up).

There’s something quite wonderful about maps too. Maybe this is another side of my geekiness but I love looking at a good map. So, what if you could have a map of a ride you’re particularly proud of turned into a piece of art to display on your wall?

As it happens – you can do exactly that. Cyced produce high quality Strava Art for runners and cyclists. So, when Angus from Cyced asked me if I’d like to review their service, I was interested to give it a try.

I settled on the ride I did with my son last year to raise money for his trip to the Kandersteg International Scout Centre and I provided a Strava link (a GPX file would have been another option). Soon afterwards, Angus sent me a PDF proof to review and, a couple of days after I confirmed the edits, the final print arrived.

I was impressed by how well it was wrapped – indeed, I’ve never had an “unboxing experience” quite like this for a piece of artwork: wrapped in tissue paper; sandwiched between sheets of heavy-duty card; all inside a sturdy card envelope. It would be pretty difficult for my postie to accidentally bend this package!

The print itself is really high quality and the simple (Strava-inspired) colours look amazing – greys, whites and orange highlights. The print that Angus created for me is A4 but there are A3 and A2 options too. I’m now considering buying a second print for the “Man Cave” when I have a suitably big ride to be proud of (or maybe with last year’s London Revolution route).

If you’re looking for something a little different for your wall – a piece of art to celebrate a ride, a run, or maybe a present for a runner/cyclist friend or family member, I’d recommending checking out the Strava Art from Cyced. As the website says it’s “worth more than just a kudos”!

Full disclosure: Cyced provided me with an A4 print in exchange for this blog post but that doesn’t influence my review. Everything I’ve written is my true opinion – but it’s nice to have the artwork for my son to keep as a memento of our 70 miles MTBing along the Grand Union Canal towpath last summer!

Working Flexibly…

Posted on By Mark Wilson

I’m not saying I’ve reached what one of my Directors once referred to as “career peak” but, as middle age firmly sets in, I can say that work is just one of many priorities I have in my life right now. I’ve probably got at least another 20 years at work, but I needed to strike a better balance between paid work, my roles as a Father and husband, and my health.

For the last few months, I’ve been working flexibly. Specifically, in my case, I’ve been working part-time, 4 days a week. Contractually 30 hours but, given that I’m output-driven, it’s probably more like 40-45. It works well for me.

A few months ago I picked up a new podcast that was being auditioned on the WB-40 podcast feed. I listened to Pauline Yau (@PaulineYau) talking about flexible working and her new venture “The Flexible Movement” and found myself thinking “Yes! Yes, I’ve seen that! And I do that!”.

Shortly afterwards, Pauline and I had a chat and the result is Episode 3 of the Flexible Movement podcast:

In the podcast, Pauline and I chat about many things but I mention how I worked with my current employer to achieve a positive outcome. I also talk about a less favourable experience elsewhere – the key difference being that I didn’t know (or follow) the right process.

Flexible working is not just for parents. My generation increasingly finds ourselves squeezed between raising a family and caring for elderly relatives but there are many reasons why people might want to break away from the “nine to five”. In the UK, employees with at least 26 weeks’ service have the right to request flexible working* once a year but employees don’t have to agree to the request. There are specific grounds though and I’m sure every company with an HR department will have a process.

The video below is produced by ACAS, and looks at some of the benefts for employers as well as some practical implications of the right to request flexible working:

And I found the following links really useful when I was requesting the change in my working practices:

“Employees that feel more in control of their work-life balance are better motivated and focused in the workplace”

Working Families, 2016

Hopefully, one day, flexible working will become “normal”. For now, it relies on a supportive culture (something this BBC article touches on). It may take a generation but I really believe one day we’ll look back office workers attending a fixed place of work for set hours on weekdays the way we look back at working for one employer our whole lives today.

*Flexible working is not “working from home on a Friday”, as some might like to think.

Caching OneDrive for Business content when Files On-Demand is enabled

Posted on By Mark Wilson

Not surprisingly, given who I work for, I’m a heavy user of Microsoft technologies. I have a Microsoft Surface Pro, running the latest versions of Windows 10 of Office 365 ProPlus, joined to Azure Active Directory and managed with Intune. I use all of the Office 365 Productivity apps. I AM A MICROSOFT POWER USER!

Enough of the drama! Let’s bring this down a level…

…I’m just a guy, using a laptop, trying to get a job done. It’s a tool.

OneDrive icon

Most of my files are stored in OneDrive for Business. There’s lots more space there than the typical SSD has available and so Microsoft introduced a feature called Files On-Demand, whereby you see the whole list of files but it’s only actually downloaded when you try to access it.

That sounds great, unless you travel a lot and work on trains and other places where network connectivity is less than ideal.

In my case, I have around 50GB of data in OneDrive and 90GB of free space on my Surface’s SSD so I have the potential to cache it all locally. I used to do this by turning off Files On-Demand but the latest build I’m running has disabled that capability for me.

It’s not feasible to touch every file and force it to be cached and I thought about asking my admins to reverse the setting to force the use of Files On-Demand but then I found another way around it…

If I right-click on a OneDrive file or folder in Windows Explorer there’s the option to “Always keep on this device”. [Update: Peter Bryant (@PJBryant) has flagged a method using the command line too – it seems there are new attributes P and U for Files On-Demand]

By applying this to one of the top-level folders in my OneDrive, I was able to force the files to be cached – regardless of whether Files On-Demand is enabled or not. Now, I can access all of the files in that folder (and any subfolders), even when I’m not connected to the Internet.

Defining multiple RADIUS servers for Aruba Wi-Fi

Posted on By Mark Wilson
Wi-Fi logo (via Pixabay)

I’ve spent some time over the last few months working with a customer who is building a complete greenfield IT infrastructure, in preparation for launching a new business. It’s been a rare privilege to work without piles of technical debt (of course, it’s never completely that simple – there is data to bring across and there are some core systems that will tie back into the parent organisation) but there have been some challenges along the way too.

One of these was when the customer’s network partner asked for a RADIUS server to be added to our identity solution (to support 802.1x based authentication for Wi-Fi clients). In itself, that wasn’t too big an ask – we could use Windows Servers running Microsoft Network Policy Server (NPS), across two Azure regions. Unfortunately, we also needed to provide resilience and the network partner was suggesting that they could only configure one IP address in their HP-Aruba cloud controllers. Azure Load Balancers only work within region and DNS round robin is not exactly smart, so myself and the other Consultants working on the solution were left scratching our heads.

Luckily, for me, having a reasonably large Twitter network meant I could ask for help – and the help came (thanks to @Tim_Siddle and others)!

We were able to take the information about server groups to our networking partner, who advised us that the cloud controllers lacked the server groups capability until recently (it was only a feature on physical controllers) but that it had now been added.

Other people responded to say they had had similar issues in the past, so this might be useful for others who are trying to configure a certificate-based authentication solution for Wi-Fi with Microsoft NPS servers.

Further reading

Enabling RADIUS Server Authentication [Aruba]