Blog

Solving the Need for Speed in Cloud Data Centers with Etched Facet Technology   Aug. 08, 2017

EFT Blog Image 2.jpg

Solving the Need for Speed in Cloud Data Centers with Etched Facet Technology 

The demand by Cloud Data Centers for faster data delivery speeds at cost-effective prices is growing rapidly. Higher speeds are necessary for Cloud Data Centers to process the current explosion in traffic. Mobile and video use drove data center traffic to 4.7 zettabytes (ZB), or almost five trillion gigabytes (GB), in 2015, and Cloud traffic will hit 14.1 ZB in 2020. Cisco predicts that in 2020, 92 percent of Data Center bandwidth will be used by cloud data workloads rather than traditional workloads. As expected, the requirements Cloud Data Centers have for resiliency and data redundancy when processing cloud data traffic necessitate high data transfer speeds.

The Need for Speed - but who is behind the wheel?

This need for speed is driven primarily by the large public cloud providers, such as Microsoft, Google, Facebook and Amazon. Analysts predict that the market for Data Center solutions will increase from an $18.6 billion market in 2015 to a $32.3 billion market in 2020. Solutions that also reduce operational cost through increased efficiency will be the most in demand.

Due to this enormous growth in traffic, Cloud Data Center operators are working to increase data delivery speeds from 100G to 400G as quickly as possible. But, there are considerations that must be made by these operators when evaluating solutions that can increase speed. These solutions will only work if they also reduce space requirements, power consumption and operational costs.

Solutions?

So, if space, power consumption and operational costs are a factor, the components (Lasers, Drivers, Amplifiers, PHYs etc.) going into these Cloud Data Centers must optimize these traits. One solution to the need for speed is the use of Etched Facet Technology (EFT) over traditional Cleaved Facet Technology (CFT). By utilizing EFT over CFT, manufacturers can offer smaller, more efficient and less expensive optical transceiver modules for optical interconnects used for high-speed data transfer at Cloud Data Centers.

EFT versus CFT

MACOM EFT.jpg

Figure 1. Source: Etched Facet Technology: Lighting the Path to 400G and Beyond in the Cloud Data Center

Let’s look at some of the reasons why EFT offers an effective solution to the need for speed. Lasers formed with the traditional CFT process are created from mechanically separated (or cleaved) wafers, and then stacked to create mirrors. EFT creates mirrors on the surface of the wafer using high-precision chemical etching. Thus, there are many advantages to using EFT in manufacturing over CFT:

  • EFT is a more precise process than CFT, therefore reducing the risk of defects due to cleaving.
  • Testing when using CFT is expensive and can only be done at laser bar level. With EFT, testing can easily be done at a full wafer level, covering the whole range of temperatures.
  • EFT provides an accurate location of the facet relative to alignment marks, whereas CFT cannot control the exact location of the facet. Using EFT, MACOM developed the proprietary self-alignment (SAEFT™) technology to attach lasers to silicon chips rather than using the slow active alignment process — thereby enabling more cost effectiveness at mass production levels.
  • EFT allows high yield production of short cavity lasers, which is critical for high speed applications.
  • EFT allows placing two lasers with a predetermined cavity length difference to increase the single mode yield.
  • EFT can reduce laser beam divergence to increase fiber coupling efficiency.
  • EFT can produce very low reflectivity facet by etching the facet at an angle, which is crucial for Semiconductor Optical Amplifier (SOA) applications.
  • EFT can generate surface emission by etching the facet at an angle and custom angles, which is critical for Silicon Photonic (SiPh) chips with grating coupler.
  • Non-hermetic passivation coating can be more effectively applied to EFT lasers on wafer scale, as compared to CFT lasers, which are coated on bar level.

Implementing the Advantages of Etched Facet Technology (EFT)

The advantages of EFT allow facets to be defined through high precision photolithography, rather than imprecise cleaving. This results in an unprecedented uniformity and yield, as well as the capability to build structures that are impossible to realize with conventional techniques. With no dependency on the crystallographic plane of the wafer, unique anti-reflection geometries can be used in place of expensive coatings.

Due to the EFT process, MACOM is able to more fully evaluate all of our lasers on the wafer in an automated, high-throughput test operation, in addition to dramatically reducing the cost of chip-handling. This process has enabled the integration of silicon PICs with lasers, creating the very first L-PIC™, or lasers integrated with a photonic integrated circuit. MACOM’s L-PIC enables a lower cost, efficient and high-yield data transfer solution, increasing the feasibility of adopting PICs as high density optical interconnect solutions for Cloud Data Centers.

In cost-sensitive markets, solutions like MACOM’s EFT process enable high precision chemical etching, minimize the risk of laser defects, reduce the overall cost of manufacturing lasers and offer a cost-effective and scalable solution, thereby delivering the products and solutions that will meet this upward need for speed. 


Read Full Post
Top 10 Trends You Need to Know About for Keeping Up with the Cloud Data Center Market   Jul. 25, 2017

DCTrends.jpg (802301300)Technology quite literally moves at the speed of light. To keep up with the Cloud Data Center market, it’s vital to stay on top of changes. Here are the top 10 trends you need to know about.

1. On-Demand Access

Cloud Data Centers primarily store information and provide disaster recovery capabilities. However, as new technologies grow—such as mobile applications and the Internet of Things (IoT)—the need for on-demand access is also growing. Users expect to have the same experience with data whether they are accessing it from a local device storage or from the cloud. Cloud Data Centers need to provide faster data processing and continue to shift focus to cloud computing and reducing latency.

2. Hiring Demand

Data scientists use analytics to turn big data into valuable and actionable insights. As Cloud Data Centers make the shift from information storage facilities to on-demand cloud data processing centers, the need for data engineers is rising. Data engineers optimize the performance of their company’s big data ecosystem. Rather than hiring just data scientists, Cloud Data Centers may need to consider hiring teams of data scientists and data engineers to create and deploy models and algorithms.

3. Infrastructure Flexibility

As many companies chase the latest technology-enabled innovation, the demand for flexible IT infrastructure grows. Many organizations are moving their data from on-premise servers to service provider Cloud Data Centers. This increases infrastructure flexibility, as businesses can choose between dedicated or shared servers, public or private clouds and hybrid services for the best fit to their rapidly shifting needs.

4. Shift from Numbers to Capacity

Despite recent explosive growth, some analysts believe that the number of Cloud Data Centers will peak in 2017. However, even as physical space demand reaches its limits, data capacity at service provider centers continues to grow and is driving datacenter interconnects to faster speeds and higher bandwidth port densities. Service providers are running mega Cloud Data Centers as they shift to cloud offerings. Cisco predicts that by 2020, 92 percent of all workloads will be in the cloud.

5. Increased Investor Interest

Cloud Data Center growth is attracting investor interest. Although some Cloud Data Center investment opportunities are less attractive due to rising energy costs and sovereignty laws, some reports indicate that Data Center Real Estate Investment Trusts (REITs) are delivering 10 percent to 15 percent returns. Other types of investment funds are yielding single digit returns in the current monetary climate, making Data Centers an attractive investment opportunity.

6. Geographic Cloud Growth

In certain markets, such as Silicon Valley, Northern Virginia (NoVa), London and Tokyo, the shift to the cloud is happening even faster than in other parts of the world. Several major providers anticipate they will need to triple infrastructure by 2020.

7. Faster Speeds Without Higher Costs

The surge in data comes from the growth of video streaming, the IoT and mobile applications. Bandwidth demands that accompany this surge are leading Data Center operators to seek faster speeds without increasing costs. 25 Gigabit Ethernet (25GbE) offers a balanced tradeoff between performance and cost as Cloud Data Centers transition to higher speeds. 

8. Bundling and Flexibility

Another way to increase data transfer and processing speeds is to bundle links. Cloud Data Centers can achieve 100G speeds by aggregating four 25GbE links. Conversely, they can increase data handling flexibility by unbundling 100GbE links into four 25GbE links.  And, the industry is already developing 400GbE links that bundle four 56GbE links using higher-order PAM4 modulation schemes.

9. High-Density Switches

Advances in semiconductor design and configuration allow Cloud Data Center operators to reduce costs through increased power efficiency. For example, a 48-port switch can now be configured on a single chip instead of the two chips previously required. Optimized solutions such as MACOM’s L-PIC™ allow Cloud Data Center operators to increase capacity cost-effectively and efficiently. By using MACOM’s proprietary self-alignment (SAEFT™) technology, MACOM’s L-PIC enables lower cost, efficient and scalable module solutions and increases the adoption of PICs as high density optical interconnect solutions for Cloud Data Centers.

10. IT Outsourcing

One way organizations can free up space on their networks while maintaining (or increasing) bandwidth speeds is to outsource their IT to the cloud. Rather than maintaining their own complete IT infrastructure, companies can turn to cloud service providers for increased speed and flexibility at a lower operating cost.

The Cloud Data Center market is evolving rapidly. Although some opportunities are shrinking as data moves to the cloud, savvy investors can realize gains. Stay on top of these 10 trends to make your investments count.  


Disclaimer

All financial guidance projections referenced in this post were made as of the publication date or another historical date noted herein, and any references to such projections herein are not intended to reaffirm them as of any later date. MACOM undertakes no obligation to update any forward-looking statement or projection at any future date. This post may include information and projections derived from third-party sources concerning addressable market size and growth rates and similar general economic or industry data. MACOM has not independently verified any information and projections from third party sources incorporated herein. This post may also contain market statistics and industry data that are subject to uncertainty and are not necessarily reflective of market conditions. Although MACOM believes that these statistics and data are reasonable, they have been derived from third party sources and have not been independently verified by MACOM.


Read Full Post
Phased Array Antennas & The Roadmap to 5G Wireless   Jul. 11, 2017

As the market and media hype around 5G continues to grow, there is a tacit acknowledgement that we have miles to go before 5G becomes a reality. Initial industry standards for 5G aren’t expected to be ratified until Summer 2018 at the earliest, and there are many regulatory issues and a myriad of technology challenges still to be resolved before 5G is ready for mainstream commercialization.

Yet, despite these daunting challenges, the promised benefits of 5G are so profound that one can’t help but get excited about it. Improved mobile phone connectivity is just the tip of the iceberg when one considers 5G’s implications for transportation, industrial and entertainment applications, among many others. In a recent whitepaper from industry research firm IHS Markit, 5G is heralded as the catalyst that will vault mobile technology into the realm of general-purpose technologies (GPTs) that drastically alter society, à la the printing press, the steam engine and electricity.

Markets Apps 5G.png

An overview of the markets and applications impacted by 5G (image above courtesy of Nokia)

MIMO vs Massive MIMO

Achieving the promise of 5G will require, among other things, major innovations in the way basestations are architected. Today we rely on multiple input, multiple output – or MIMO – antennae configurations to multiply the capacity of antennae links for wireless basestations. These antennas provide the ability to concentrate signal strength into smaller areas of space, boosting overall efficiency and throughput by guiding the signal to the precise location it’s needed. By adding additional antennas, this beamforming capability is improved.

But whereas conventional basestations may house between two and eight antennas, 5G basestations will need anywhere from 64 to hundreds of antennas arrayed in a “massive MIMO” configuration to provide the requisite data rates. This phased array antennae design comprises an active electronically scanned array (AESA), which enables signals to be electronically steered with much greater beamforming precision than MIMO can support today.

High-Performance, Low Cost Active Antennas

When it comes to the architecture and assembly of massive MIMO 5G systems, we see many parallels with the new generation of Multifunction Phased Array Radar (MPAR) active antennae systems targeted for use for military and civil air traffic control, and weather system tracking applications. And while you might not typically associate this class of radar system with cost-sensitive commercial applications like 5G, you might be surprised to learn that MPAR technology leverages design and manufacturing efficiencies that dramatically reduce the cost of the end system.

First generation MPAR systems feature an array of Scalable Planar Array (SPAR™) Tiles in a flat panel configuration comprised of hundreds to thousands of active antennas. SPAR Tile technology, developed in a collaboration between MACOM and MIT Lincoln Laboratory, embodies a new cost-conscious approach to phased array radar system development, leveraging highly-integrated antenna sub-systems, and volume scale commercial packaging and manufacturing techniques.

 

MACOM MPAR.jpg

First generation MPAR systems leverage an array of Scalable Planar Array (SPAR) Tiles – shown above

Tile-based AESAs create the foundation for a new generation of high-performance, agile radar systems that can be built quickly and flexibly tailored and scaled for deployment across a range of applications – at 5X less cost than conventional slat array architectures. Continued innovations in phased-array-based technologies like MPAR will help to unlock the full promise of 5G, allowing basestation OEMs to simplify design and manufacturing processes, and get to market faster with 5G technology.

To learn more about the benefits of MPAR technology for 5G, check out this article in Microwave Product Digest. For a deep-dive read on the MPAR technology architecture, head over to Microwave Journal


Read Full Post
The Future of Cloud Data Center Markets and What Passive Optical Networks Can Teach Us   Jun. 27, 2017

DCPONBlog.jpg (694828330)Cloud Data Centers are growing rapidly. One analyst’s report found that major cloud providers expect to triple infrastructure by 2020. This growth is fueled by increasing bandwidth demands but comes along with the need to reduce power consumption and control costs. As Cloud Data Centers grow, Cloud Data Center networking architecture must evolve efficiently and effectively meet the demand for bandwidth capacity while maximizing throughput speeds and lowering the average cost-per-bit. 

From Copper Cable to Passive Optical Networks (PONs) 

Passive Optical Networks (PONs) represent a significant leap forward from traditional copper cable networks. PONs offer broader reach, and faster and more secure data transfers at a lower cost. Fiber optic cables used by PONs are smaller than copper cables, require less space, and are able to carry more data. This approach reduces power consumption and lowers facilities, equipment, installation, and maintenance costs. Since PONs are easier to maintain, and fiber optic cables are physically more secure than copper cable, security is enhanced, there is reduced downtime and breakage, and the ability to withstand malicious attacks is increased. 

Applying the Lessons Learned from PONs to Cloud Data Center Infrastructure 

Cloud Data Center growth is increasing at a rapid pace. In 2014, Google spent $10 billion on building new Cloud Data Centers, and another $10 billion on maintaining existing Cloud Data Centers. As reported by The Government Accountability Office, it is estimated that the United States government spends $81 billion a year across all IT expenditures, including mission critical computing systems some of which are more than half a century old. Modernizing these systems to maximize efficiency and lower costs thus represents an enormous market opportunity for manufacturers of the technologies that help Cloud Data Center operators achieve these goals.                                            

Demand for bandwidth capacity is driven by media generation and consumption trends, including photos and videos on social media; a shift to viewing Over-the-Top content; and developing augmented reality for gaming and advertising. Bandwidth demand also comes from sources such as e-commerce, the increased collection and analysis of data across industries and institutions, the growth of data sets to petabytes in size, and the growing Internet of things (IOT). It’s no longer just computers and phones, everything from televisions to tractors are now connected to the Internet. 

PONs provide a blueprint for transforming Cloud Data Center architecture into networks that are capable of handling more data at faster data speeds and lower cost.

The Next Generation of PONs and the Future of Cloud Data Centers 

Given the nearly insatiable demand for increased connectivity and higher throughput, there is a multi-billion-dollar opportunity to provide network component solutions to Cloud Data Centers. Building on current expertise in PONs, the next generation of market-leading Cloud Data Center networking technologies such as optical networking, optical transceivers, and silicon photonics with integrated self-aligning Etched Facet Technology (EFT) lasers (SAEFT™) are being developed. 

In addition to bandwidth demands, cost considerations are also driving opportunities in providing solutions to Cloud Data Centers. Cloud Data Center operators seek to maximize speed and throughput at the lowest cost-per-bit possible. Breakthroughs in optical networking target this goal. PONs demonstrated that it is possible to meet scale and cost requirements via efficient optics technologies. Cloud Data Centers can achieve their cost, speed, and efficiency goals by applying these principles. 

Similarly, EFT technology will transform optical connectivity within Cloud Data Centers, enabling the cost efficiencies and hyper scale growth agility necessary to keep pace with an estimated five trillion gigabytes of annual Cloud Data Center traffic – a number that will continue to skyrocket as millions of new users join the cloud-connected apps economy and IoT. By leveraging technologies and solutions such as MACOM’s EFT laser technology, and L-PICTM transmitters enabled by EFT and SAEFT, along with commercial scale manufacturing capability, Cloud Data Center opportunities can be capitalized on by replicating the cost structure reductions that have been achieved in Passive Optical Networks. 


Read Full Post

All Blog Posts