Public cloud adoption will continue to go up and up – but as hybrid initiatives go up with them, concerns persist over how to handle cloud migration challenges.

That’s the key finding from the latest study by VIAVI Solutions. The IT and network testing provider, in its most recent State of the Network Global Study, polled more than 600 IT professionals and found more than half (56%) of enterprises polled had made the leap to public cloud as of this year. This number is set to go up to 72% by 2020.

By 2020, more enterprises plan to run a larger percentage of their apps in the cloud. As of this year, 62% of firms polled say only up to a quarter of their apps are cloud-based. By 2020, this number is set to dip to 28%, with 44% saying they have between a quarter and half of their apps in the cloud. More than a quarter (28%) of those polled said they expected at least half of their apps to be cloud-based in two years – a number which is only at 11% today.

Yet while these figures all look impressive, it is akin to the duck analogy – serene on top but paddling like hell underneath. Two thirds (65%) of those polled said their network team was responsible for troubleshooting cloud issues, compared with 24% who said they weren’t.

As VIAVI puts it, this translates as a disconnect between IT and business; if IT are fighting fires post-migration, then the chances are they weren’t around pre-migration, as the move may have been decided by a specific business unit. More than half (52%) said their biggest problem was determining whether problems were caused by the network, the application, or the system – by some distance the most frequently cited issue.

In terms of what needs to be done, the company suggests four key takeaways; adopting a ‘cloud by default’ approach; ensure engineers have as much visibility into the organisation’s SaaS offerings as possible; get visibility for the full application journey, from the user, to the cloud and the data centre, and get remote user intelligence on board.

“Based on this year’s State of the Network, you’ve really got to feel for enterprise IT teams. They’re losing control of infrastructure and services that are migrating to the cloud, while simultaneously supporting employees who may be working anywhere, and yet remain on the hook to maintain performance and resolve issues,” said Douglas Roberts, VIAVI enterprise and cloud business unit VP and GM. “They need solutions to increase their scope of visibility and speed of response, as well as smarter analytics.”

If your organisation is not allowing its developers to use container technologies, then you will very soon be in the minority, according to the latest analysis from DigitalOcean.

The company, in its latest quarterly Currents report assessing developer trends in cloud computing, found 49% of devs are now using containers in some form. Of that number, JavaScript (57%) was the most popular language used with them, ahead of Python (46%), PHP (36%), and Go (28%). Three in five (60%) said they were using containers both for testing and development and production, while three quarters (78%) of those who aren’t using containers today still plan to adopt them.

Scalability is the name of the game for container adoption according to the report, with 39% of respondents citing it as the key aspect. Simpler software testing (24%), quicker software testing (23%) and avoiding vendor lock-in (10%) were also cited.

Not altogether surprisingly, Kubernetes is the biggest game in town. 42% of those polled say they use it, compared with 35% for nearest rival Docker Swarm. Red Hat’s OpenShift (5%), Apache Mesos (3%), and CoreOS Tectonic (1%) all polled negligibly. Yet smaller companies are more likely to be Docker houses – of those with five employees or fewer, Docker Swarm (41%) won out over Kubernetes (31%).

Compared with a relatively strong consensus on where container technologies sit, developers’ knowledge and enthusiasm over serverless computing was somewhat split. Only half of those polled said they had a strong understanding of it, with four in five of those who don’t (81%) saying they plan to do further research this year. Around one in three – 35% for the US, 32% for the UK – say they have deployed applications in a serverless environment over the past year, with AWS Lambda (58%) the most popular platform, ahead of Google Cloud Functions (23%).

With this in mind, what should prospective employees and their employers be looking for? 39% of devs polled said their top considerations for new jobs were a competitive salary and opportunity for internal growth, while the company product (17%) and freedom to use particular technologies (23%) were lower down on the list. The research also found that many developers are still going down the traditional college route – more than half (51%) said they went to college, compared with only 6% who attended a coding bootcamp.

The report polled more than 4,800 respondents, with more than half (55%) saying they were developers, 13% working in DevOps and 10% saying they were students or managers respectively.

You can read the full report here.

Alibaba Cloud’s expansion outside of China and Asia Pacific has been well documented – and now the provider is firming up its commitments with the launch of a new EMEA partner program.

The partner program will aim to look at four key areas; developing digital transformation in targeted vertical industries; supporting the development of talent; advancing technology innovation; and enhancing marketplaces.

The overall effect, in the company’s words, is ‘to create an inclusive ecosystem that can benefit all those involved’. What this means is the support of companies such as Intel and Accenture, as well as Station F, the world’s largest startup campus.

“Our goal in EMEA is to bring powerful and elastic cloud services to our customers and create a well-connected, comprehensive ecosystem with our partners to accelerate cloud technology development in the regional cloud industry,” said Yeming Wang, Alibaba Cloud EMEA general manager in a statement.

Speaking to this publication back in May, Wang noted the changing landscape as key to Alibaba Cloud’s proposed expansion. Not only is it a global strategy from the whole Alibaba group, but there are political and technological ramifications – from China opening its doors to outside trade more on the one hand, to a rise in multi-cloud initiatives on the other.

“Today, we have a lot of clients asking to adopt Alibaba as a second or third public cloud provider,” said Wang. “Alibaba Cloud, from a user experience point of view, is quite similar to AWS. That is why we got comments from different clients – they say ‘if the guys are AWS certified or [an] AWS expert, then you’re halfway to being very familiar with Alibaba also.’”

Amazon is certainly the target for Alibaba globally. According to the latest note from analyst firm Synergy Research, AWS leads across all geographies, with Microsoft and Google comprising the top three everywhere instead of Asia Pacific, where Alibaba is second.

We've all heard the exhortations from vendors and industry players alike – get into DevOps, containers and serverless for IT and application delivery before it's too late. But while the benefits are evident, companies are getting bogged down in troubleshooting and debugging software issues.

That's according to a new report from Scalyr. The study, which polled 155 software development practitioners, found that the majority (71%) are getting used to DevOps-type practices, pushing code into production at least weekly. Almost a third of respondents said they did so at least once a day.

Yet while greater efficiencies are coming out at one end, observability and analysis shoots up at the other. Almost half of those polled said they have five or more observability tools, with 58% of respondents in a DevOps role doing so. When it comes to log management, the biggest concern for software developers is ad-hoc query speed; this was cited by 54% of overall respondents, with that number going up to 68% for those who mostly work in microservices.

Two in five (40%) said their companies' engineers spent the majority of their time troubleshooting software issues, with the number rising to almost three quarters (73%) of companies who push out code on a daily basis. A quarter said they spent the majority of their log management investigation time waiting for queries to complete.

The figures show that time saved in one area is essentially relative – developers are still wasting a lot of time doing something else. "As organisations make the shift from more traditional architectures to microservices and deliver code more frequently, they spend more time troubleshooting and debugging software, working to understand complex data sources, and operational visibility is paramount," wrote Jamie Barnett, Scalyr chief customer officer in a blog post.

"Our takeaway is that organisations that are experiencing this shift to modern software delivery need to take a hard look at their observability tools and processes to make sure they can keep pace and evolve to support modern, high-speed, distributed software engineering practices," Barnett added.

This chimes in with other recent research on the topic. According to the Ponemon Institute last month, the gap between what organisations ideally want in DevOps and microservices practices versus what they are actually able to deliver is costing them millions each year.

IBM is looking to build upon recent cloud momentum – and the company is expanding in Europe after securing several new customers in healthcare, logistics, energy and more.

The announcements showcase how many prospective IBM clients are utilising the company’s cloud for its artificial intelligence, machine learning and blockchain capabilities. Credit Mutuel, a French bank, is deploying IBM Watson virtual assistants across all of its business lines – run on IBM’s cloud in France with a backup in Germany – while Koopman Logistics, based in the Netherlands, will aim to track and trace consignments across its supply chain through IBM’s blockchain.

Alongside them are Gruppo 24 Ore, a media firm based in Italy, Spanish digital health provider Teckel Medical, UK-based RS Components and lighting solutions firm Osram AG, based in Germany.

Last month at CeBIT, IBM announced 18 new availability zones across the North America, Europe and Asia Pacific regions, among other launches designed around security and privacy. “Our new availability zones and regions architecture is the next step in the evolution of our public cloud platform, and it’ll immediately reinforce and supplement the broad portfolio of infrastructure, platform, and software services that our clients trust to fuel their businesses,” wrote Andrew Hately, VP, DE and chief architect IBM Watson and Cloud Platform at the time.

Speaking to this publication back in February, John Considine, IBM general manager of cloud infrastructure services, cited the importance of extracting data to glean actionable insights for businesses as key – with the emerging technologies forming part of these extraction methods.

“One of our theories leading into the cloud, for the past few years, is that data is enormously important for the enterprises – and given more than 80% of the world’s data is still maintained behind the corporate firewall, our focus has been how… we enable the businesses to take advantage of that data, to combine it with new processing techniques, new data sets, and new capabilities,” Considine said.

“[It’s about] all the things associated with machine learning and deep learning, analytics and bringing all of these things together in a form that allows them to tap into those resources and deliver not only application modernisation, but really even process reinvention,” he added.

It is important to note, as Considine did, how much data is in less-than-easy spots. IBM punted out a similar statistic – that almost 80% of all enterprise data is still managed on the mainframe – when a new partnership with CA Technologies was announced last month.

Google Cloud Platform has said it will conduct a detailed review of its abuse prevention processes after a customer complained about its treatment.

The unnamed customer, who works in the renewable energy industry, wrote in a Medium post that the company was a few days away from ‘losing everything’ after Google’s automated system pinged it for questionable activity.

Those who get pinged will receive a variety of emails – a ‘barrage’, as the customer put it – detailing that each service is down, the payments account is temporarily closed, and what needs to be done about it. Chat support is switched off, with a warning that unless a picture of the credit card and a government-issued photo ID of the card holder is uploaded within three days, the project will be deleted.

The customer warned about the consequences if the card holder – in this instance, the CFO – was not available, and around the automated nature of the system.

“I understand Google’s need to monitor and prevent suspicious activity. But how you handle things after some suspicious activity is detected matters a lot,” the post explains. “You need a human element here – one that cannot be replaced by any amount of code/AI. You just can’t turn things off and then ask for an explanation.”

In a statement, posted by Brian Bender, Google Cloud Platform engineering support regional lead, Google said it will be re-evaluating data sources used to assess potential fraudulent activity, implementing additional mechanisms for suspect accounts, and improve how it communicates account warnings. “Protecting our customers and systems are a top priority,” the statement added. “We sincerely apologise for this issue and are working quickly to make things better, not just for this customer but for all GCP customers.”

Given Google’s rise in the cloud infrastructure arena over the past 12 months – the company was listed in the leaders’ section for public cloud IaaS by Gartner in May – it is interesting to note that, for this particular customer, this was the first project built by them entirely on Google’s cloud. The customer was previously an AWS house; and while there was no technical reason cited for the change – both are ‘on-par’, as the customer put it – there was a note on the differing customer experiences.

“In our experience AWS handles billing issues in a much more humane way,” the customer explained. “They warn you about suspicious activity and give you time to explain and sort things out. They don’t kick you down the stairs.”

Other issues were sorted. Mike Kahn, Google Cloud customer engineer, noted the importance of having an enterprise user account rather than a consumer one – yet another commenter described this approach as having 'borderline contempt' for customers.

It has been described by Citrix as a ‘never-ending technology arms race to optimise performance’ – and now Formula 1 has gotten a further boost by selecting Amazon Web Services (AWS) as its official cloud and machine learning provider.

The move will see Formula 1 move the vast majority of its infrastructure from on-premises data centres to AWS, and use a variety of products to help improve broadcasts, data tracking, and race strategies.

Amazon SageMaker – AWS’ service to help developers build, train and deploy machine learning models – will be put to task by Formula 1’s team of data scientists against more than 65 years of race data. The data, collected in real time by Amazon Kinesis and stored in Amazon DynamoDB and cold storage product Glacier, will be crunched to extract performance statistics and make predictions for upcoming races.

Other AWS products being utilised by Formula 1 are AWS Lambda for serverless capabilities, and AWS Elemental Media Services for greater video options.

“For our needs, AWS outperforms all other cloud providers, in speed, scalability, reliability, global reach, partner community, and breadth and depth of cloud services available,” said Pete Samara, Formula 1 director of innovation and digital technology in a statement. “By leveraging Amazon SageMaker and AWS’s machine learning services, we are now able to deliver these powerful insights and predictions to fans in real time.”

This is by no means the first customer to cite machine learning as a key element of future strategies. In May, Ryanair announced it was going all-in on AWS, saying greater data insights and better customer experience through machine learning was vital to its decision. The airline is using Amazon Lex, the technology underpinning smart assistant Alexa, on a trial basis.

Software delivery and automation provider Puppet has raised $42 million (£31.9m) in series F funding to complement its recent expansion.

The round, which was led by Cisco Investments, alongside EDBI, Kleiner Perkins, True Ventures and VMware, comes only a few weeks after Puppet announced the acquisition of Reflect, a company which provides data visualisation as a service. Total funding for the company now stands at $149.5 million across seven rounds in total.

Puppet certainly has reasons to be bullish about its scope. According to the company’s 2017 DevOps report, 66% of DevOps engineers and 69% of software engineers polled in the US get paid more than $100,000 per year, with the overarching message being that enterprises are getting to grips with new ways of delivering IT services and software.

It is evidently this route of giving users insights to deliver better services – whether it’s through data visualisation, as in the Reflect acquisition, or otherwise – driving Puppet’s recent gains. The company cited Gartner figures from earlier this year which argue that by 2020, 90% of the top 100 global companies will be using DevOps practices to significantly cut operational inefficiencies.

“Our rapid growth and international expansion is a testament to the rising demand for DevOps transformation, software automation and the pressing need for enterprises to navigate the new world of software delivery,” said Sanjay Mirchandani, CEO of Puppet. “That’s why we’ve been so focused on expanding our product portfolio – to empower customers to discover, deliver and operate software across their cloud and containerised environments.”

The company opened five new offices over the past year, in Seattle, Singapore, Sydney, Timisoara and Tokyo. When this publication focused on Puppet in May last year, the first three were in the pipeline, with the company saying there was ‘global momentum’ in its results.

2017 was a ‘pivotal’ year in expanding public cloud service adoption according to IDC – with spending growth remaining at a constant level despite the overall market tripling in size.

The figures, which come from the analyst firm’s latest Worldwide Semiannual Public Cloud Services Tracker, show that while the overall growth rate for 2017 was a little smaller than the previous year, revenue growth of the top 16 providers by market share went up. The top tier vendors now capture more than half (50.7%) of the overall market.

Software as a service (SaaS) remains the largest bucket by some distance at $74.8 billion (£56.5bn) globally, with IDC predicting the SaaS market will hit $163bn by 2022 – well ahead of overall 2017 figures of $116.7bn. Infrastructure as a service (IaaS) spending last year was at $24.9bn, while platform as a service (PaaS) was at $17bn.

PaaS remained the fastest growing of the three markets, showing a 47.1% year on year increase, compared with IaaS (39.9%) and SaaS (22.4%). Both PaaS and IaaS however saw slightly slower growth when compared to the previous year’s figures, of 48% and 45% respectively.

Breaking down the numbers into regional figures, IDC found the US continues to provide the bulk of public cloud services revenue, although again seeing a minor decline from 62% to 60%. This is a trend which will continue in the coming years with new regional services and expansion from global players cited.

These figures make for interesting reading when compared with analysis from Synergy Research also published late last week. Synergy focused more on geographical dispersion, finding that aside from Asia Pacific, Amazon Web Services (AWS) led the way ahead of Microsoft and Google. The anomaly is down to the rise of China – where the top five providers are all local companies – with Alibaba second in APAC and fourth worldwide. Synergy described public cloud as ‘essentially a global market’ that was ‘a game of scale… and to be a market leader demands vast ongoing investments, a global presence and a global brand.’

“2017 saw some intriguing market share shifts among the major players, as all of them have significantly increased their focus on the cloud, and competitive pressure has ratcheted way up,” said Frank Gens, IDC senior vice president and chief analyst. “The next three years will determine IT industry leadership for the next two decades and beyond.”

Amazon Web Services (AWS) has total geographic dominance in public cloud – but while Microsoft and Google secure second and third place through most of the globe, Alibaba is a clear second in APAC.

This is the key finding from the latest note by Synergy Research, which focused on Q1 data – although the majority of the figures won’t differ from what many market watchers already recognise.

AWS ranks at #1 worldwide, ahead of Microsoft, Google, Alibaba, and IBM. Across North America and EMEA, the top three remain the same but with IBM and Salesforce taking the last two places, while in Latin America the latter two swap over. In APAC, AWS leads Alibaba, ahead of Microsoft and Google, with Tencent taking fifth position – the latter because of its strong market placing in China.

Indeed, in China the top five providers are all local companies – and it is this sign which, as regular readers of this publication will recognise, has enabled Alibaba to expand further afield. As Yeming Wang, general manager of Alibaba Cloud Europe, explained last month: “To go global is definitely a corporate level of strategy.”

“Despite some local data sovereignty and regulatory issues, in most meaningful ways public cloud computing is essentially a global market,” said John Dinsdale, a chief analyst and research director at Synergy Research. “This is a game of scale and to be a market leader demands vast ongoing investments, a global presence and a global brand.

“Of course there will often be local issues that might enable local companies to carve out niche positions for themselves, but they will remain small local players in a specific country or sub-region,” added Dinsdale. “It is also true that in such cases the global leaders can usually deploy different local strategies to enable them to succeed.

“With the glaring exception of China, we view this as a truly global market.”