Tuesday, September 27th, 2016

Time	Event
12:00p	When Selecting a Data Center Provider, Keep an Eye on the Edge As your business and user base become more distributed, there’s a lot more data to keep an eye on. You need to deploy technologies that can keep pace with today’s digital demands. This means working with the right kind of data center partners, service providers who can keep up with your business. Today, organizations across all verticals and sizes are looking for better ways to control and distribute their data. Many have turned to cloud. Many others are now looking at deeper cloud services to differentiate. A lot of companies are working hard to create better architectures around data distribution and content delivery networks. They are actively working to optimize data delivery. In October 2000, IDC estimated that the US content distribution market will increase from $10 million in 1999 to nearly $1 billion in 2004. A recent MarketsandMarkets report showed that the global CDN market size is expected to grow from $4.95 Billion in 2015 to $15.73 Billion in 2020. That’s $10 Million in 1999, and $15 Billion by 2020. This rate of growth indicates that many SMBs and enterprises are trying to control data and how it’s being delivered. A good data center partner can help your business grow; an even better partner can help your business evolve. This is why many organizations are looking for data center and cloud partners who can take their data to the edge, and beyond. Read more: How Edge Data Center Providers are Changing the Internet’s Geography Keeping an Eye on Data at the Edge In working with a data center provider, it’s important to create good SLAs, make sure they can support your business and grow with the overall market trends. It’s also important — especially if you’re a growing business with numerous locations and distributed users — to ensure that your partner can properly manage all of your data. With that in mind, here are a few things to look out for when working with a data center partner while focusing on keeping your data agile and at the edge: Ethernet services, data center interconnects, and cloud providers. Your data center isn’t a closed building. In fact, a good data center partner provides interconnects, good Ethernet services, and a way to actually integrate with other cloud providers. That said, some cloud and data center providers already offer some type of CDN services: Amazon, Azure, Internap, Level 3, Rackspace, and several others. Furthermore, your telco can help you with CDN as well. BT Group, AT&T, NTT, Telefonica, Telus, and Verizon all offer ways to create CDN capabilities. You can also create your own CDN platform by leveraging systems like OnApp, Aryaka, CloudFlare, and others. However, you need to work with a data center provider to ensure your data can be distributed economically and cached effectively. Just being able to can send data across into another data center isn’t enough. Work with your data center provider to ensure that CDNs are (or at least can be) built into the contract. Looking into edge and CDN capabilities. As mentioned earlier, there are already data center providers who can offer very powerful CDN options. Amazon CloudFront, for example, has servers located in Europe (United Kingdom, Ireland, The Netherlands, Germany, Spain), Asia (Hong Kong, Singapore, Japan, Taiwan and India), Australia, South America, as well as in several major cities in the US. The service operates from 55 edge locations on five continents. Similarly, Microsoft Azure CDN has 38 points of presence worldwide. From there, you can create your own management capabilities. When working with a CDN provider or partner – make sure you can control the data that’s flowing in and out. In some cases, native CDN management might work. If that’s not enough, look for technologies which offer easy management integration via methods like REST APIs, for example. Similarly, you’ll also find powerful management options with open source code available in C#/.NET, Python, PHP, Java, Ruby, and others. Creating optimization around heavy data distribution. If your data is heavily distributed, don’t work with a data center partner that has only one or two locations or a fledgling CDN offering. It’s critical for you to deliver the data as efficiently as possible while still keeping costs down. Remember, CDNs are designed to help businesses operate better and to improve overall user experiences. Furthermore, it all heavily depends on the type of content you’re trying to deliver. For example, platforms like Yottaa are designed for web and mobile data services optimization. Others, like Aryaka specifically aim to create powerful WAN optimization solutions aimed at data delivery and distribution. The point here is that technologies around data delivery and CDN offerings has grown. Depending on the level of distribution, the size of your organization, and your overall requirements – selecting the right data center provider will always revolve around three specific points: your users, your business, and your data. If you have a global organization, work with a data center provider ready to take on massive data distribution requirements. However, if you’re a smaller or mid-market organization working in specific regions, work with a data center provider that can deliver data to your markets. Your data center(s) house your critical applications, user environments, and business tools. Today’s IT ecosystem has become the lifeblood of business. Moving forward, there will be more growth around data, the requirement to deliver this data economically and efficiently, and more requirements around QoS/QoE technologies. Good data center partners don’t just offer solid internal services and IT services. They look for ways they can bridge your infrastructure with other cloud services. If your partner is too stringent and closed off from other cloud offerings and partners, you might need to look for an alternative, fast. Those data center providers which embrace their place as data distribution points will be the ones who help win in this very competitive digital market. (Comment on this)
6:53p	Latest Microsoft Data Center Design Gets Close to Unity PUE Microsoft says it is now close to achieving unity PUE, or perfect infrastructure energy efficiency, with its latest data center design. Innovation in data center design by cloud giants has widespread implications for their users as well as for the data center industry in general. Improving data center efficiency is one of the main levers cloud providers can pull to lower the price of their services. As they innovate in pursuit of lower and lower cost at greater and greater scale, many of those innovations eventually get adopted by companies designing data centers for other purposes. Using a mix of data centers it builds and operates on its own and leased facilities, Microsoft’s cloud now lives in more than 100 data centers in 34 regions around the world. The company is now on its fifth-generation data center design. In a rare look at the physical backend of its global cloud infrastructure, the company has produced a video tour (registration required) of its data center campus in Quincy, Washington, which showcases three design generations, including the latest one. The fifth-generation facility is a culmination of everything the company’s infrastructure team has learned about data center design over the previous generations. “We’ve learned something from every step of the way,” Christian Belady, general manager of cloud infrastructure strategy and architecture at Microsoft, said in the video. “We’ve adopted it and brought it into this design.” The facility’s PUE (Power Usage Effectiveness) is down to 1.1, and even below 1.1 at certain times of the year, he said. It’s “almost at unity, so it’s a major step.” While PUE has been championed by The Green Grid, the data center industry organization responsible for making it the most widely used data center energy efficiency metric around the world, Belady is the one who created the concept about a decade ago while designing a data center for a customer. PUE measures the power overhead of electrical and mechanical infrastructure by comparing total power that comes into a data center with the amount of power used by the IT equipment inside. Microsoft doesn’t appear to have come up with any groundbreaking innovations to achieve its extremely low PUE. The biggest differences between the latest design and the previous-generation one are a move away from using data center containers, or ITPACs, as the company called them, and a switch to waterside economization from a mix of outside air and adiabatic cooling that was employed in ITPACs. Microsoft spokesman Alex Brady with ITPACs in the background on Microsoft’s Quincy, Washington, data center campus (Source: Microsoft video) As we reported earlier this year, the company has stopped using ITPACs because the model couldn’t support the rate of global data center capacity expansion it has been undertaking as it grows its cloud services business, namely Azure and Office 365. The fifth-gen Microsoft data center design is a building where racks that ship to the site already packed with pre-tested servers are rolled in from the loading dock along a bare concrete floor. The lack of raised floor is a departure from Microsoft’s pre-ITPAC days, when it used traditional raised-floor-based cooling design. The new design uses a closed-loop waterside economization system. Water is cooled in massive cooling towers outside of the facility and pushed through the cooling loop. Inside the building, the loop goes through a heat exchanger, from where a wall of fans pushed cold air into the IT hall. Fan wall inside Microsoft’s fifth-generation data center in Quincy, Washington (Source: Microsoft video) The design also employs a few reliable efficiency best practices, such as hot-aisle containment and optimal inlet air temperature. Because Microsoft designs its own servers, it has been a priority to ensure they can run at relatively high temperatures, which reduces the demand for cooling capacity. Belady did not specify inlet air temperature inside the fifth-generation Microsoft data center, but servers inside its ITPAC-based gen-four data centers run without any air conditioning when outside air temperature is around 80F. Once outside air reaches above 90F or so, the cooling surface of the adiabatic coolers gets wetted to bring the temperature down, he said. In general, the latest Microsoft data center design looks close to what Facebook and Google have been doing in their facilities. The overall theme is simplification and industrialization. These facilities look closer to manufacturing plants than to the lab-like raised-floor IT environments you will find inside the more traditional enterprise data centers. “It feels very industrial,” Belady said. “Everything’s very simple.” (Comment on this)
7:17p	Hospital Pays $400,000 HIPAA Breach Penalty for Obsolete ‘Business Associate’ Agreement Brought to you by MSPmentor A Rhode Island hospital agreed this month to pay $550,000 in settlements after failing to properly update business associate agreements as required under the privacy and security rules of the Health Insurance Portability and Accountability Act (HIPAA), federal authorities said. The U.S. Department of Health and Human Services Office of Civil Rights (OCR) opened an investigation into Women & Infants Hospital of Rhode Island (WIH) after receiving a report of a data breach in November 2012. WIH told federal authorities it had lost unencrypted backup tapes containing ultrasounds of 14,004 women, including patient names, dates of birth, dates of exams, physician names and, in some cases, Social Security numbers. Information technology services, including information security, were handled by WIH’s parent company, Care New England Health Systems (CNE). “WIH provided OCR with a business associate agreement with Care New England Health System effective March 15, 2005, that was not updated until Aug. 28, 2015, as a result of OCR’s investigation, and therefore, did not incorporate revisions required under the HIPAA Omnibus Final Rule,” according to a Sept. 23 OCR news release announcing the settlements. See also: HIPAA Breach Case Results in Record $5.5M Penalty IT Services Provider Pays $650,000 HIPAA Breach Fine The total amount to be paid by WIH is actually comprised of two settlements. A $400,000 payment is intended to address the federal probe, which found that WIH disclosed protected health information (PHI) to CNE, without “obtaining satisfactory assurances as required under HIPAA,” in the form of a written business associate agreement that CNE would safeguard the PHI. “This case illustrates the vital importance of reviewing and updating, as necessary, business associate agreements, especially in light of required revisions under the Omnibus Final Rule,” said OCR Director Jocelyn Samuels. “The Omnibus Final Rule outlined necessary changes to established business associate agreements and new requirements which include provisions for reporting,” she continued. “A sample Business Associate Agreement can be found on OCR’s website to assist covered entities in complying with this requirement.” Another $150,000 consent judgment is being paid to the Massachusetts Attorney General’s Office in response to the hospital’s conduct in the underlying breach, including failing to provide adequate safeguards and failing to notify affected people in a timely manner. “While the AGO’s actions do not legally preclude OCR from imposing civil money penalties, OCR determined not to include additional potential violations in this case for the purposes of settlement, given that such potential violations had already been addressed by the AGO and based on OCR’s policy approach to concurrent cases with State AGOs,” the federal news release said. The $400,000 settlement with OCR brings the total amount of settlements for HIPAA security violations to $20.7 million this year, up sharply from $6.2 million in all of 2015. This first ran at http://mspmentor.net/msp-mentor/hospital-pays-400000-hipaa-breach-penalty-obsolete-business-associate-agreement (Comment on this)
7:24p	Cisco Said to Spend Up to $4B to Lift Mexico Production (Bloomberg) — Cisco Systems Inc. plans to spend as much as $4 billion in Mexico over the next couple of years, upgrading its factories and increasing production through contract manufacturers, a person with knowledge of the matter said. The biggest maker of equipment that runs the internet, based in San Jose, California, will announce the plans Tuesday, said the person, who asked not to be identified because the news isn’t public yet. The spending figure, which may not be publicly disclosed with the announcement, includes contracts with manufacturers, as opposed to money spent directly on factories, and includes spending that had already been planned. A Cisco spokesman declined to comment. Routing is Cisco’s second-largest revenue source. The plan is a boost to Mexican President Enrique Pena Nieto as he seeks to show his economic reforms are attracting more investment. The timing could be delicate for Cisco — six weeks after the company announced job cuts and a day after the U.S. presidential debate, in which Republican candidate Donald Trump decried the flow of jobs to Mexico and other countries, threatening to increase taxes on companies that move investment outside the U.S. It would also be one of the first major investment announcements by a U.S. company in Mexico since April, when Ford Motor Co. said it would spend $1.6 billion on a new small-car factory in Mexico, drawing a rebuke from Trump. In June of last year AT&T Inc. said it would invest about $3 billion to extend mobile Internet service to Mexico in addition to spending $4.4 billion earlier in the year to acquire Iusacell and Nextel Mexico. Cisco said in August it plans to cut about 7 percent of its workforce. Savings from the job reductions will be invested in newer businesses that Cisco expects to fuel sales growth, such as cloud computing and connected services, the company said. (Comment on this)

<< Previous Day

2016/09/27 [Calendar]

Next Day >>