Microsoft Azure VM Types Comparison

Microsoft Azure VM Types Comparison

Microsoft Azure VM types come in a wide range optimized to meet various needs. Machine types are specialized, and vary by virtual CPU (vCPU), disk capability, and memory size, offering a number of options to match any workload.

With so many options available, finding the right machine type for your workload becomes confusing – which is why we’ve created this overview of Azure VM types (as we did before with EC2 instance types, and Google Cloud machine types). Note that while AWS EC2 instance types have names associated with their purpose, Azure instance type names are simply in a series from A to N.The chart below and written descriptions are a brief and easy reference, but remember that finding the right machine type for your workload will always depend on your needs.

General Purpose

General purpose VMs are suitable for balanced CPU and memory, making them a great option for testing and development, smaller to medium databases, and web servers with lower traffic:

DC-series

The latest family of virtual machines stand out for data protection and code confidentiality. SGX technology and a 3.7GHz Intel XEON E-2176G Processor back these machines, and in conjunction with Intel Turbo Boost Technology, they can go up to 4.7 GHz.

Av2 Series

A-series VMs have a CPU-to-memory ratio that works best for entry level workloads, like those for development and testing. Sizing is throttled for consistent processor performance to run the instance.

Dv2-series

Dv2 VMs boast powerful CPUs – roughly 35% faster than D-series VMs – and optimized memory, great for for production workloads. With the same memory and disk configurations as the D-series, based upon either a 2.4 GHz or 2.3 GHz processor and Intel Boost Technology, they can go to up to 3.1 GHz.

Dv3-series

With expanded memory and adjustments for disk and network limits, the Dv3 series Azure VM type offers the most value to general purpose workloads. Best for enterprise applications, relational databases, in-memory caching, and analytics.

B-series

Similar to the AWS t-series machine type family, B-series VMs are burstable and ideal for workloads that do not rely on full and continuous CPU performance. Customers can purchase a VM size that builds up credits when underutilized, and the accumulated credits can be used as bursts – spikes in compute power that allow for higher CPU performance when needed. Use cases for B-series VM types include development and testing, low-traffic web servers, small databases, micro services, and more.

Dsv3-series

With premium storage and a 2.4 or 2.3 GHz Intel Xeon processor that can achieve 3.5 GHz thanks to Intel Turbo Boost Technology 2.0, the Dsv3-series is best suited for most production workloads.  

Compute Optimized

Compute optimized Azure VM types offer a high CPU-to-memory ratio. They’re suitable for medium traffic web servers, network appliances, batch processing, and application servers.

Fsv2-series

With a base core frequency of 2.7 GHz and a maximum single-core turbo frequency of 3.7 GHz, Fsv2 series VM types offer up to twice the performance boost for vector processing workloads. Not only do they offer great speed for any workload, the Fsv2 also offers the best value for its price based on the ratio of Azure Compute Unit (ACU) per vCPU.

F-series

F-series Azure VM types are great for workloads that require speed thanks to the 2.4 GHz Intel Xeon processor, reaching speeds up to 3.1 GHz with the Intel Turbo Boost Technology 2.0. The F-series is your best bet for fast CPUs but not so much when it comes to memory or temporary storage per vCPU. Analytics, gaming servers, web servers, and batch processing would work well with the F-series.

Memory Optimized

Memory optimized VM types are higher in memory as opposed to CPU, and best suited for relational database services, analytics, and larger caches.

M-Series

Enterprise applications and large databases will benefit most from the M-series for having the most memory (up to 3.8 TiB) and the highest vCPU count (up to 128) of any VM in the cloud.

Dv2-series, G-series, and the DSv2/GS

For applications that require fast vCPUs, reliable temporary storage, and demand more memory, the Dv2, G, and DSv2/GS series all fit the bill for enterprise applications. The Dv2 series offers a speed and power with a CPU about 34% faster than that of the D-series. Based on the 2.3 and 2.4 GHz Intel Xeon® processors and with Intel Turbo Boost Technology 2.0, they can reach up to 3.1 GHz. The Dv2-series also has the same memory and disk configurations as the D-series.

Ev3-series

The Ev3 follows in the footsteps of the high memory VM sizes originating from the D/Dv2 families. This Azure VM types provides excellent value for general purpose workloads, boasting expanded memory (from 7 GiB/vCPU to 8 GiB/vCPU) with adjustments to disk and network limits per core basis in alignment with the move to hyperthreading.

Storage Optimized

For big data, SQL, and NoSQL databases, storage optimized VMs are the best type for their high disk throughput and IO.

Ls-series

VMs provide as much as 32 vCPUs with the Intel® Xeon® processor E5 v3 family. The Ls-series comes with the same CPU performance as the G/GS-Series and 8 GiB of memory per vCPU. This type works best applications requiring low latency, high throughput, and large local disk storage.

GPU

GPU VM types, specialized with single or multiple NVIDIA GPUs, work best for video editing and heavy graphics rendering – as in compute-intensive, graphics-intensive, and visualization workloads.

  • NC, NCv2, NCv3, and ND sizes are optimized for compute-intensive and network-intensive applications and algorithms.
  • NV and NVv2 sizes were made and optimized for remote visualization, streaming, gaming, encoding, and VDI scenarios.]

High Performance Compute

For the fastest and most powerful virtual machines, high performance compute is the best choice with optional high-throughput network interfaces (RDMA).

H-series

For the latest in high performance computing, the H-series Azure VM was built for handling batch workloads, analytics, molecular modeling, and fluid dynamics. These 8 and 16 vCPU VMs are built on the Intel Haswell E5-2667 V3 processor technology featuring DDR4 memory and SSD-based temporary storage.

And besides sizable CPU power, the H-series provides options for low latency RDMA networking with FDR InfiniBand and different memory configurations for supporting memory intensive compute requirements.

What Azure VM type is right for you?

With six virtual machine types belonging to multiple families and coming in a range of sizes, how do you determine the right Azure VM type for your workload? The good news is that with this many options, you’re bound to find the right type to meet your computing needs – as long as you know what those needs are. With good insight into your workload, usage trends, and business needs, you’ll be able to find the Azure VM type that’s right for you.

 

Even if you’re not (yet) multi-cloud, you should use cloud agnostic tools

Even if you’re not (yet) multi-cloud, you should use cloud agnostic tools

There’s a simple fact for public cloud users today: you need to use cloud agnostic tools. Yes – even if you only use one public cloud. Why? This recommendation comes down to a few drivers that we see time and time again.

You won’t always use just this cloud

There is an enterprise IT trend to multi-cloud and hybrid cloud – such a prevalent trend that even if you are currently single-cloud, you should plan for the eventuality of using more than one cloud, as the multi-cloud future has arrived. Dave Bartoletti, VP and Principal Analyst at Forrester Research, who broke down multi-cloud and hybrid cloud by the numbers:

  • 62 percent of public cloud adopters are using 2+ unique cloud platforms
  • 74 per cent of enterprises describe their strategy as hybrid/multi-cloud today

In addition, standardizing on cloud agnostic tools also can alleviate costs associated with policy design, deployment, and enforcement across different cloud environments. Management and monitoring using the same service platform greatly reduces the issue of mismatched security policies and uncertainty in enforcement. Cloud agnostic tools that also operate in the context of the data center — whether in a cloud, virtualized, container, or traditional infrastructure — are a boon for organizations who need to be agile and move quickly. Being able to reuse policies and services across the entire multi-cloud spectrum reduces friction in the deployment process and offers assurances in consistency of performance and security.

How do you decide what tools to adopt?

We talk to different size enterprises using the cloud on a daily basis, and always ask if they are using cloud native tools, or if they are using third party tools that are cloud agnostic. The answer – it’s a mix to be sure, often it’s a mix between cloud-native and third-party tools within the same enterprise.

What we hear is that managing the cloud infrastructure is quite a complex job, especially when you have different clouds, technologies, and a diverse and opinionated user community to support. So a common theme with many of the third-party tools we see used tend to include freemium models, a technology someone used at a previous company, tools recommended by the cloud services provider (CSP) themselves, and open-API-driven solutions that allow for maximum automation in their cloud operations. It also serves the tools vendors well if deploying the tool includes minimum effort — in other words, SaaS tools that do not require a bunch of services and integration work. Plug and play is a must.

For context, here at ParkMyCloud support AWS, Azure, Google and Alibaba clouds, and usually talk to DevOps and IT Ops folks responsible for their cloud infrastructure. And those folks are usually after cloud cost control and governance when speaking with us. So our conversations tend to focus on the tools they use and need for cloud infrastructure management like CI/CD, monitoring, cost control, cost visibility and optimization, and user governance. For user governance and internal communication, Single-sign On and ChatOps are must have.

So we decided to compile a list of the most common clouds and tools we run across here at ParkMyCloud, in order of popularity:

  • Cloud Service Provider
    • AWS, Google Cloud, Microsoft Azure, Alibaba Cloud – and we do get requests for IBM and Oracle clouds
  • Infrastructure Monitoring (not APM)
    • Cloud Native (AWS CloudWatch, Azure Metrics, Google Stackdriver), DataDog, Nagios, SolarWinds, Microsoft, BMC, Zabbix, IBM
  • Cost Visibility and Optimization
    • CloudHealth Technologies, Cloudability, Cloudyn/Azure Cost Management, Apptio
  • CI/CD + DevOps (this is broad but these are most common names we hear that fit into this category)
    • Cloud Native, CloudBees Jenkins, Atlassian Bamboo, HashiCorp, Spinnaker, Travis CI
  • Single Sign-On (SSO)
    • ADFS, Ping, Okta, Azure AD, Centrify, One Login, Google OAuth, JumpCloud
  • ChatOps
    • Slack, Microsoft Teams, Google Hangouts
  • Cloud Cost Control
    • Cloud Native/Scripter, ParkMyCloud, GorillaStack, Skeddly, Nutanix (BotMetric)

Beat the curve with cloud agnostic tools

Our suggestion is to use cloud agnostic tools wherever possible. Our experience tells us that a majority of the enterprises lean this way anyways. The upfront cost in terms of license fee and/or set up could be more, but we think it comes down to (1) most people will end up hybrid/multi-cloud in the future, even if they aren’t now, and (2) cloud agnostic tools are more likely to meet your needs as a user, as the companies building those tools will stay laser-focused on supporting and improving said functionality across the big CSPs.

Google Hangouts & Microsoft Teams Integrations for Cloud Server Monitoring

New in ParkMyCloud: we’ve released integrations with chat clients Google Hangouts and Microsoft Teams to make cloud server monitoring easier and integrated into your day. Now, ParkMyCloud users can get notifications when their resources are about to turn on or off, when a user overrides a schedule, and more.

We created these integrations based on popular demand! ParkMyCloud has had a Slack integration since last summer. Now, we’re encountering more and more teams that set themselves up as pure Google or pure Microsoft shops, hence the need. If your team only uses Google tools – Google Cloud Platform for cloud, Google OAuth for SSO, and Google Hangouts for chat — you can use ParkMyCloud with all of these. Same with Microsoft: ParkMyCloud integrates with Microsoft Azure, ADFS, and Microsoft Teams.  

ParkMyCloud notifications in Google Hangouts – note the “view resource” link will take you straight to the resource in ParkMyCloud

Here’s what actions ParkMyCloud admins can get notified on through a chat client for better cloud server monitoring:

  • Resource Shutdown Warning – Provides a 15-minute warning before an instance is scheduled to be parked due to a schedule or expiring schedule override.
  • User Actions – These are actions performed by users in ParkMyCloud such as manual resource state toggles, attachment or detachment of schedules, credential updates, etc.
  • Parking Actions – These are actions specifically related to parking such as automatic starting or stopping of resources based on defined parking schedules.
  • Policy Actions – These are actions specifically related to configured policies in ParkMyCloud such as automatic schedule attachments based on a set rule.
  • System Errors – These are errors occurring within the system itself such as discovery errors, parking errors, invalid credential permissions, etc.
  • System Maintenance and Updates – These are the notifications provided via the banner at the top of the dashboard.

There are a few ways these can be useful. If you’re an IT administrator and you see your users toggling resource states frequently, the notifications may help you determine the best parking schedule for the users’ needs.

Or let’s say you’re a developer deep in a project and you get a notification that your instance is about to be shut down — but you still need that instance while you finish your work. Right in your Microsoft Teams window, you can send an override command to ParkMyCloud to keep the instance running for a couple more hours.

ParkMyCloud notifications in Microsoft Teams

These integrations give ParkMyCloud users better perspective into cloud server monitoring, right in the same workspaces they’re using every day. Feedback? Comment below or shoot us an email – we are happy to hear from you!

P.S. We also just created a user community on Slack! Feel free to join here for cloud cost, automation, and DevOps discussions.

Cloud User Management Comparison: AWS vs. Azure vs. GCP vs. Alibaba Cloud

Cloud User Management Comparison: AWS vs. Azure vs. GCP vs. Alibaba Cloud

When companies move from on-prem workloads to the cloud, common concerns arise around costs, security, and cloud user management. Each cloud provider handles user permissions in a slightly different way, with varying terminology and roles available to assign to each of your end users. Let’s explore a few of the differences in users and roles within Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform, and Alibaba Cloud.

AWS IAM Users and Roles

AWS captures all user and role management within IAM, which stands for “Identity and Access Management”. Through IAM, you can manage your users and roles, along with all the permissions and visibility those users and service accounts have within your AWS account. There are a couple different IAM entities:

  • Users – used when an actual human will be logging in
  • Roles – used when service accounts or scripts will be interacting with resources

Both users and roles can have IAM policies attached, which give specific permissions to operate or view any of the other AWS services.

Azure RBAC

Azure utilizes the RBAC system within Resource Manager for user permissions, which stands for “Role Based Access Control”. Granting access to Azure resources starts with creating a Security Principal, which can be one of 3 types:

  • User – a person who exists in Azure Active Directory
  • Group – a collection of users in Azure Active Directory
  • Service Principal – an application or service that needs to access a resource

Each Security Principal can be assigned a Role Definition, which is a collection of permissions that they can utilize to view or access resources in Azure. There are a few built-in Role Definitions, such as Owner, Contributor, Reader, and User Access Administrator, but you can also create custom role definitions as well depending on your cloud user management needs.  Roles may be assigned on a subscription by subscription basis.

Google Cloud Platform IAM

Google Cloud Platform also uses the term IAM for their user permissions. The general workflow is to grant each “identity” a role that applies to each resource within a project. An identity can be any of the following:

  • Google account – any user with an email that is associated with a Google account
  • Service account – an application that logs in through the Google Cloud API
  • Google group – a collection of Google accounts and service accounts
  • G Suite domain – all Google accounts under a domain in G Suite
  • Cloud Identity domain – all Google accounts in a non-G-Suite organization

Roles in Google Cloud IAM are a collection of permissions. There are some primitive roles (Owner, Editor, and Viewer), some predefined roles, and the ability to create custom roles with specific permissions through an IAM policy.

Alibaba Cloud RAM

Alibaba Cloud has a service called RAM (Resource Access Management) for managing user identities. These identities work in slightly different ways than the other cloud service providers, though they have similar names:

  • RAM-User – a single real identity, usually a person but can also be a service account
  • RAM-Role – a virtual identity that can be assigned to multiple real identities

RAM users and roles can have one or more authorization policies attached to them, which in turn can each have multiple permissions in each policy. These permissions then work similarly to other CSPs, where a User or Role can have access to view or act upon a given resource.

Cloud User Management – Principles to Follow, No Matter the Provider

As you can see, each cloud service provider has a way to enable users to access the resources they need in a limited scope, though each method is slightly different. Your organization will need to come up with the policies and roles you want your users to have, which is a balancing act between allowing users to do their jobs and not letting them break the bank (or your infrastructure). The good news is that you will certainly have the tools available to provide granular access control for your cloud user management, regardless of the cloud (or clouds) you’re using.

4 Cloud Computing Jobs to Check Out if You Want to Break Into the Space

4 Cloud Computing Jobs to Check Out if You Want to Break Into the Space

Lately, we’ve been thinking about cloud computing jobs and titles we’ve been seeing in the space. One of the great things about talking with ParkMyCloud users is that we get to talk to a variety of different people. That’s right – even though we’re laser-focused on cloud cost optimization, it turns out that can matter to a lot of different people in an organization. (And no wonder, given the size of wasted spend – that hits people’s’ buttons).

You know the cloud computing market is growing. You know that means new employment opportunities, and new niches in which to make yourself valuable. So what cloud computing jobs should you check out?

If you are a sysadmin or ops engineer:

Cloud Operations. Cloud operations engineers, managers, and similar are the people we speak with most often at ParkMyCloud, and they are typically the cloud infrastructure experts in the organization. This is a great opportunity for sysadmins looking to work in newer technology.

If you’re interested in cloud operations, definitely work on certifications from AWS, Azure, Google, or your cloud provider of choice. Attend meetups and subscribe to industry blogs – the cloud providers innovate at a rapid pace, and the better you keep up with their products and solutions, the more competitive you’ll be.

See also: DevOps, cloud infrastructure, cloud architecture, and IT Operations.

If you like technology but you also like working with people:

Customer Success, cloud support, or other customer-facing job at a managed service provider (MSP). As we recently discussed, there’s a growing market of small IT providers focusing on hybrid cloud in the managed services space. The opportunities at MSPs aren’t limited to customer success, of course – just in the past week we’ve talked to people with the following titles at MSPs: Cloud Analyst, Cloud Engineer, Cloud Champion/Cloud Optimization Engineer, CTO, and Engagement Architect.

Also consider: pre-sales engineering at one of the many software providers in the cloud space.

If you love process:

Site Reliability Engineer. This title, invented by Google, is used for operations specialists who focus on keeping the lights on and the sites running. Job descriptions in this discipline tend to focus on people and processes rather than around the specific infrastructure or tools.  

If you have a financial background:

Cloud Financial Analyst. See also: cloud cost analyst, cloud financial administrator, IT billing analyst, and similar. Cloud computing jobs aren’t just for technical people — there is a growing field that allows experts to adapt financial skills to this hot market. As mentioned above, since the cloud cost problem is only going to grow, IT organizations need professionals in financial roles focused on cloud. Certifications from cloud providers can be a great way to stand out.

What cloud computing jobs are coming next?

As the cloud market continues to grow and change, there will be new cloud computing job opportunities – and it can be difficult to predict what’s coming next. Just a few years ago, it was rare to meet someone running an entire cloud enablement team, but that’s becoming the norm at larger, tech-forward organizations. We also see a trend of companies narrowing in “DevOps” roles to have professionals focused on “CloudOps” specifically — as well as variations such as DevFinOps. And although some people hear “automation” and worry that their jobs will disappear, there will always be a need for someone to keep the automation engines running and optimized. We’ll be here.

Multi-Cloud, Hybrid Cloud, and Cloud Spend – Statistics on Cloud Computing

Multi-Cloud, Hybrid Cloud, and Cloud Spend – Statistics on Cloud Computing

The latest statistics on cloud computing all point to multi-cloud and hybrid cloud as the reality for most companies. This is confirmed by what we see in our customers’ environments, as well as by what industry experts and analysts report. At last week’s CloudHealth Connect18 in Boston we heard from Dave Bartoletti, VP and Principal Analyst at Forrester Research, who broke down multi-cloud and hybrid cloud by the numbers:

  • 62% of public cloud adopters are using 2+ unique cloud environments/platforms
  • 74% of enterprises describe their strategy as hybrid/multi-cloud today
  • But only:
    • 42% regularly optimize cloud spending
    • 41% maintain an approved service catalog
    • 37% enforce capacity limits or expirations

More often than not, public cloud users and enterprises have adopted a multi-cloud or hybrid cloud strategy to meet their cloud computing needs. Taking advantage of features and capabilities from different cloud providers can be a great way to get the most out of the benefits that cloud services can offer, but if not used optimally, these strategies can also result in wasted time, money, and computing capacity.

The data is telling – but we won’t stop there. For more insight on the rise of multi-cloud and hybrid cloud strategies, and to demonstrate the impact on cloud spend (and waste) – we have compiled a few more statistics on cloud computing.

Multi-Cloud and Hybrid Cloud Adoption Statistics

The statistics on cloud computing show that companies not only use multiple clouds today, but they have plans to expand multi- and hybrid cloud use in the future:

  • According to a 451 Research survey, 69% of organizations plan to run a multi-cloud environment by 2019. As they said, “the future of IT is multi-cloud and hybrid” – but with this rise, cloud spending optimization also becomes more of a challenge.
  • In a survey of nearly 1,000 tech executives and cloud practitioners, over 80% of companies were utilizing a multi-cloud strategy, commonly including a hybrid cloud model consisting of both public and private clouds.
  • And by multi-cloud, we don’t mean just two. On average, the number of private and public clouds used by companies to run applications and test out new services is 4.8.
  • On hybrid cloud strategy:
    • 83% of workloads are virtualized today (IDC)
    • 60% of large enterprises run VMs in the public cloud (IDC)
    • 65% of organizations have a hybrid cloud strategy today (IDC)

Cloud Spend Statistics

As enterprises’ cloud footprints expand, so too does their spending:

  • It’s not just public – the rise in cloud spend is happening on all fronts. According to IDC, 62.3 percent of private cloud spending went to on-premise private clouds in 2017.
  • The increase in cloud use, along with the rise of multi-cloud and hybrid cloud strategies, also correlates with an increased investment in cloud services. In a survey of nearly 1,000 tech executives and cloud practitioners, 20% of enterprises plan to increase their cloud spend by more than double, and another 17% plan to up their cloud spending by 50-100%, according to the report.  
  • 75% of participants said that one of their primary concerns was the challenge of managing cloud spend. Cloud cost optimization was a priority for the majority of participants, and average cloud waste was reported at 35%.
  • In another study from 451 Research, 38.8% of CIOs said that “cost savings” was their biggest motivator in migrating to the cloud, but post migration, cloud costs was the biggest challenge they faced. Here’s what else they had to say:

“Cloud is an inexpensive and easily accessible technology. People consume more, thereby spending more, and forget to control or limit their consumption. With ease of access, inevitably some resources get orphaned with no ownership; these continue to incur costs. Some resources are overprovisioned to provide extra capacity as a ‘just in case’ solution. Unexpected line items, such as bandwidth, are consumed. The IT department has limited visibility or control of these items.”

What Does ParkMyCloud User Data Tell Us?

We’ve noticed some interesting patterns in the cloud platforms adopted by ParkMyCloud users as well, which highlight the multi-cloud trends discussed above as well as correlations between the types of companies that are attracted to each of the major public clouds. We observed:

  • A high rate of growth in the number of Google Cloud Platform (GCP) customers over the past several months. While Amazon Web Services still holds the lion’s share among organizations using ParkMyCloud, the rate of growth is much higher for GCP. We believe that as more and larger organizations become enmeshed in GCP’s infrastructure, they are finding a greater need for cost optimization.
  • Among our customers using a multi-cloud strategy, the majority use AWS in combination with Azure, while the rest are using AWS with Google Cloud Platform.
  • The adoption model for AWS and GCP users are similar – both allow small to medium business to start small and adopt quickly, while still supporting the largest companies in the world. On the other hand, Azure customers tend to adopt through Enterprise License Agreements. We encounter few startups using Azure.

What These Statistics on Cloud Computing Mean for Cloud Management  

Upon examining these statistics on cloud computing, it’s clear that multi-cloud and hybrid cloud approaches are not just the future, they’re the current state of affairs. While this offers plenty of advantages to organizations looking to benefit from different cloud capabilities, using more than one CSP complicates governance, cost optimization, and cloud management further as native CSP tools are not multi-cloud. As cloud costs remain a primary concern, it’s crucial for organizations to stay ahead with insight into cloud usage trends to manage spend (and prevent waste). To keep costs in check for a multi-cloud or hybrid cloud environment, optimization tools that can track usage and spend across different cloud providers are a CIO’s best friend.