No, Ericsson says.
Tuesday, February 19, 2019
Is Edge Computing Just Hype?
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Wi-Fi Offload Might Grow in Early Years of 5G
The amount of traffic offloaded from smartphones will be 59 percent by 2022, Cisco predicts.
The amount of traffic offloaded from 4G was 57 percent at the end of 2017, and it will be 59 percent by 2022, Cisco estimates. On 5G networks, Wi-Fi offload could be as high as 71 percent, Cisco estimates.
The amount of traffic offloaded from 3G will be 40 percent by 2022, and the amount of traffic offloaded from 2G will be 30 percent.
“As 5G is being introduced, while we expect plans to be generous with data caps and speeds will be higher than ever, the new application demands on 5G are also going to move upwards as well encouraging similar behaviors of offload as 4G,” Cisco believes.
Also, the fixed network will continue to supply most of the actual consumed bandwidth, as much as 80 percent, in 2022, Cisco estimates.
The offload percentage on 5G is estimated to be 71 percent by 2022. However, as the 5G network matures, we may see offload rates come down, Cisco says.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
IoT Usage Climbs, Vodafone Finds
The latest Vodafone IoT Barometer found 34 percent of respondent companies are now using Internet of Things applications, up from 29 percent in the previous IoT Barometer.
Regionally, the Americas region saw the biggest increase, where adoption rose from 27 percent to 40 percent. The industries that saw the greatest increase were transport and logistics (27 percent to 42 percent) and manufacturing and industrials (30 percent to 39 percent).
Different IoT projects have different needs. For example, an application that monitors the storage conditions of medical samples doesn’t need to send lots of data, but reliability is critical. One which tracks the location of individual parcels doesn’t need to send updates every second, but it needs to be cost-effective.
Edge computing use cases are like any others: they hinge on value and cost. Sometimes the key value is ultra-low latency; sometimes it is the cost of moving data around; in other cases it is the dependability of communications.
Some 72 percent of respondents said that they have a project where data must be delivered within seconds, or less. But 39 percent said that they have an application where a transmission delay of hours wouldn’t matter. And 21 percent said they have both.
The amount of data devices need to transmit can also vary greatly between projects. Some 70 percent of users said that they have an IoT project that sends a lot of data, more than 10 MB per device per day, while 39 percent said that they have an IoT project where devices only send a small amount of data. About 20 percent said that they have both.
Sometimes keeping costs down is crucial to making an IoT business case work. Some 43 percent of respondents said that they have an application where they would trade slower delivery of messages for a reduction in cost. Conversely, 64 percent said that they’d pay more for consistently fast delivery of messages. About 25 percent said that both were true.
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Monday, February 18, 2019
Has Japan Internet Access Speed Fallen, or Not, Compared to Other Countries?
By one analysis conducted by Nikkei Asian Review, typical end user internet access speed in Japan has fallen, in a comparative sense, since 2015, though Japan’s average speed ranking, compared to other OECD countries, has risen since 2010.
But the broadband speed data seems inconsistent. Other studies do now show that shift in speed ranking. Organization for Economic Cooperation and Development statistics, for example, still show Japan in the top ranks for typical internet download speed.
Those sorts of differences in average speeds and country ranking are common. To be sure, changes happen all the time.
The Nikkei study suggests Japan's evening communication speeds are slower than those of Thailand and Malaysia which shored up their optical communication lines. Taiwan and Singapore have surpassed Japan for average daytime speed.
According to that analysis, Tokyo's speed was about 12.6 Mbps in January-April 2018. Sweden and Denmark speeds are now around 40 Mbps, while the United States and United Kingdom have overtaken Japan in communications speeds.
Nikkei extracted and analyzed data from 199 countries and regions, aided by the University of Tokyo's Koshizuka Laboratory. The analysis was conducted using data released by M-Lab, a communications speed measurement project involving U.S. tech company Google and Princeton University.
Japan's communications speed varies significantly between day and night, when more videos are watched. The country's typical speed of 5 Mbps at 10 p.m. is just one-fourth of its morning figure. The average nighttime speed is comparable to that of Russia's, Nikkei argues.
Korea 28.6
Norway 23.5
Sweden 22.5
Switzerland 21.7
Finland 20.5
Japan 20.2
Denmark 20.1
United States 18.7
Netherlands 17.4
Czech Republic 16.9
United Kingdom 16.9
Latvia 16.6
Belgium 16.3
Canada 16.2
Ireland 15.6
Spain 15.5
Germany 15.3
OECD average 15.25
Hungary 14.8
New Zealand 14.7
Lithuania 14.6
Austria 14.1
Israel 13.7
Slovenia 13.7
Slovak Republic 13
Portugal 12.9
Poland 12.6
Russia 11.8
Estonia 11.6
Luxembourg 11.6
Australia 11.1
France 10.8
Chile 9.3
Italy 9.2
Greece 7.9
Turkey 7.6
Mexico 7.5
Colombia 5.5
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Living on the Edge
I have been writing this blog for more than a decade. Over that time, I found I was doing enough work specifically on mobility and wireless to add Spectrum Matters and Spectrum Futures, looking at mobility/wireless on the former site, and spectrum issues on the latter site.
With the emergence of 5G, it seems clear enough to me that other topics, including core network virtualization, ultra-low latency use cases and edge computing are going to be intertwined.
In that regard, I am adding a new site dealing with "infrastructure edge" computing, defined as edge computing capabilities operated by connectivity providers. The reason is that enterprise use cases and revenues enabled by infrastructure edge computing are ultimately going to emerge as the key new revenue opportunity 5G enables.
While 5G is quite useful for consumer mobility (representing the next platform to boost bandwidth and download speeds), I do not see big new industry revenue opportunities emerging from 5G in the consumer arena.
Most of that will come from use cases related broadly to internet of things, and most of that will come from the enterprise applications. From a connectivity provider perspective, it seems obvious enough that much of the important new industry revenue upside comes from roles as suppliers of the edge computing functions.
So Living on the Edge is where you will find content on infrastructure edge computing (edge computing capabilities supplied by mobile and fixed service providers).
With the emergence of 5G, it seems clear enough to me that other topics, including core network virtualization, ultra-low latency use cases and edge computing are going to be intertwined.
In that regard, I am adding a new site dealing with "infrastructure edge" computing, defined as edge computing capabilities operated by connectivity providers. The reason is that enterprise use cases and revenues enabled by infrastructure edge computing are ultimately going to emerge as the key new revenue opportunity 5G enables.
While 5G is quite useful for consumer mobility (representing the next platform to boost bandwidth and download speeds), I do not see big new industry revenue opportunities emerging from 5G in the consumer arena.
Most of that will come from use cases related broadly to internet of things, and most of that will come from the enterprise applications. From a connectivity provider perspective, it seems obvious enough that much of the important new industry revenue upside comes from roles as suppliers of the edge computing functions.
So Living on the Edge is where you will find content on infrastructure edge computing (edge computing capabilities supplied by mobile and fixed service providers).
Gary Kim has been a digital infra analyst and journalist for more than 30 years, covering the business impact of technology, pre- and post-internet. He sees a similar evolution coming with AI. General-purpose technologies do not come along very often, but when they do, they change life, economies and industries.
Subscribe to:
Posts (Atom)
-
We have all repeatedly seen comparisons of equity value of hyperscale app providers compared to the value of connectivity providers, which s...
-
It really is surprising how often a Pareto distribution--the “80/20 rule--appears in business life, or in life, generally. Basically, the...
-
Who gets to use spectrum, and concerns about interference from other users, now appears to be an issue for Google’s Project Loon in India. ...
